Skip to main content
NCBI home page
BMJ Clinical Evidence logoLink to BMJ Clinical Evidence
. 2010 Feb 11;2010:0604.

Obesity in adults

David DeLaet 1,#, Daniel Schauer 2,#
PMCID: PMC2907622

Abstract

Introduction

About one third of the US population and one quarter of the UK population are obese, with increased risks of hypertension, dyslipidaemia, diabetes, cardiovascular disease, osteoarthritis, and some cancers. Fewer than 10% of overweight or obese adults aged 40 to 49 years revert to a normal body weight after 4 years. Nearly 5 million US adults used prescription weight-loss medication between 1996 and 1998, but one quarter of all users were not overweight.

Methods and outcomes

We conducted a systematic review and aimed to answer the following clinical questions: What are the effects of drug treatments in adults with obesity? What are the effects of bariatric surgery in adults with morbid obesity? We searched: Medline, Embase, The Cochrane Library, and other important databases up to May 2009 (Clinical Evidence reviews are updated periodically; please check our website for the most up-to-date version of this review). We included harms alerts from relevant organisations such as the US Food and Drug Administration (FDA) and the UK Medicines and Healthcare products Regulatory Agency (MHRA).

Results

We found 34 systematic reviews, RCTs, or observational studies that met our inclusion criteria. We performed a GRADE evaluation of the quality of evidence for interventions.

Conclusions

In this systematic review, we present information relating to the effectiveness and safety of the following interventions: bariatric surgery versus medical interventions; biliopancreatic diversion; diethylpropion; gastric bypass; gastric banding; mazindol; orlistat (alone and in combination with sibutramine); phentermine; rimonabant; sibutramine (alone and in combination with orlistat); sleeve gastrectomy; and vertical banded gastroplasty.

Key Points

About one third of the US population and one quarter of the UK population are obese, with increased risks of hypertension, dyslipidaemia, diabetes, CVD, osteoarthritis, and some cancers.

  • Fewer than 10% of overweight or obese adults aged 40 to 49 years revert to a normal body weight after 4 years.

  • Nearly 5 million US adults used prescription weight-loss medication between 1996 and 1998, but one quarter of all users were not overweight.

Orlistat, phentermine, rimonabant, and sibutramine may promote modest weight loss (an additional 1-7 kg lost) compared with placebo in obese adults having lifestyle interventions, but they can all cause adverse effects.

  • Sibutramine may be more effective at promoting weight loss compared with orlistat, although not in obese people with type 2 diabetes or hypertension.

  • We don't know whether combining orlistat and sibutramine treatment leads to greater weight loss than with either treatment alone.

  • We don't know whether diethylpropion and mazindol are effective at promoting weight loss in people with obesity.

  • Orlistat has been associated with GI adverse effects.

  • Phentermine has been associated with heart and lung problems.

  • Sibutramine has been associated with cardiac arrhythmias and cardiac arrest. In January 2010, the European Medicines Agency suspended marketing authorisation of sibutramine in the European Union because of the increased risk of non-fatal myocardial infarctions and strokes.

  • Rimonabant has been associated with an increased risk of psychiatric disorders.

Bariatric surgery (gastric bypass, vertical banded gastroplasty, biliopancreatic diversion, or gastric banding) may increase weight loss compared with no surgery in people with morbid obesity.

Compared with each other, we don't know whether gastric bypass, vertical banded gastroplasty, biliopancreatic diversion, or gastric banding is the most effective surgery or the least harmful.

  • We don't know whether sleeve gastrectomy is effective.

  • Bariatric surgery may result in loss of over 20% of body weight, which may be largely maintained for 10 years.

  • Operative and postoperative complications are common, and on average 0.28% of people die within 30 days of surgery. Mortality may be as high as 2% in some high-risk populations. However, surgery may reduce long-term mortality compared with no surgery.

About this condition

Definition

Obesity is a chronic condition characterised by an excess of body fat. It is most often defined by the BMI, a mathematical formula that is highly correlated with body fat. BMI is weight in kilograms divided by height in metres squared (kg/m2). Worldwide, adults with a BMI of 25 to 30 kg/m2 are categorised as overweight, and those with a BMI above 30 kg/m2 are categorised as obese. Nearly 5 million US adults used prescription weight-loss medication between 1996 and 1998. One quarter of users were not overweight. Inappropriate use of prescription medication is more common among women, white people, and Hispanic people. The National Institutes of Health (NIH) in the USA has issued guidelines for obesity treatment, which indicate that all obese adults (BMI more than 30 kg/m2), and all adults with a BMI of 27 kg/m2 or more and with obesity-associated chronic diseases are candidates for drug treatment. Morbidly obese adults (BMI more than 40 kg/m2), and all adults with a BMI of 35 kg/m2 or more and with obesity-associated chronic diseases are candidates for bariatric surgery.

Incidence/ Prevalence

Obesity has increased steadily in many countries since 1900. In the UK in 2002, it was estimated that 23% of men and 25% of women were obese. In the past decade alone, the prevalence of obesity in the USA has increased from 22.9% between 1988 and 1994, to 34% in 2006.

Aetiology/ Risk factors

Obesity is the result of long-term mismatches in energy balance, where daily energy intake exceeds daily energy expenditure. Energy balance is modulated by a myriad of factors, including metabolic rate, appetite, diet, and physical activity. Although these factors are influenced by genetic traits, the increase in obesity prevalence in the past few decades cannot be explained by changes in the human gene pool, and it is more often attributed to environmental changes that promote excessive food intake and discourage physical activity. Less commonly, obesity may also be induced by drugs (e.g., high-dose glucocorticoids, antipsychotics, antidepressants, oral hypoglycaemic agents, and antiepileptic drugs), or be secondary to various neuroendocrine disorders, such as Cushing's syndrome and PCOS.

Prognosis

Obesity is a risk factor for several chronic diseases, including hypertension, dyslipidaemia, diabetes, CVD, sleep apnoea, osteoarthritis, and some cancers. The relationship between increasing body weight and mortality is curvilinear, where mortality is highest among adults with very low body weight (BMI less than 18.5 kg/m2) and among adults with the highest body weight (BMI more than 35 kg/m2). Obese adults have more annual admissions to hospitals, more outpatient visits, higher prescription drug costs, and worse health-related quality of life than normal-weight adults. Fewer than 10% of overweight or obese adults aged 40 to 49 years revert to a normal body weight after 4 years.

Aims of intervention

To achieve realistic gradual weight loss, and prevent the morbidity and mortality associated with obesity, without undue adverse effects.

Outcomes

Reduction in mortality; adverse effects of treatment. We found no RCTs that assessed the primary outcome of reduction in mortality associated with obesity. Proxy measures assessed in studies included mean weight loss (kg), proportion of people losing 5% or more of baseline body weight, and proportion of people maintaining weight loss.

Methods

Clinical Evidence search and appraisal May 2009. The following databases were used to identify studies for this review: Medline 1966 to May 2009, Embase 1980 to May 2009, and The Cochrane Database of Systematic Reviews and Cochrane Central Register of Controlled Clinical Trials, Issue 2, 2009 (1966 to date of issue). We also searched for retractions of studies included in the review. Abstracts of the studies retrieved from the initial search were assessed by an information specialist. Selected studies were then sent to the contributor for additional assessment, using predetermined criteria to identify relevant studies. Study design criteria for inclusion in this review were: published systematic reviews and RCTs in any language. RCTs for drug interventions had to be at least single blinded, for surgical interventions open or blinded studies were acceptable. Studies had to contain 20 or more people. We have excluded RCTs assessing drug treatments with less than 4 months' follow-up, and RCTs assessing surgical treatments with less than 1 year's follow-up. We included systematic reviews of RCTs and RCTs where harms of an included intervention were studied applying the same study design criteria for inclusion as we did of benefits. We have also excluded RCTs with greater than 30% loss to follow-up unless they performed an intention-to-treat analysis. However, such RCTs may be included in the meta-analyses of systematic reviews. We did not perform a search for observational studies of bariatric surgery. However, we have included all observational studies of bariatric surgery identified by systematic reviews. In addition, we use a regular surveillance protocol to capture harms alerts from organisations such as the US Food and Drug Administration (FDA) and the UK Medicines and Healthcare products Regulatory Agency (MHRA), which are added to the reviews as required. To aid readability of the numerical data in our reviews, we round many percentages to the nearest whole number. Readers should be aware of this when relating percentages to summary statistics such as relative risks (RRs) and odds ratios (ORs). We have performed a GRADE evaluation of the quality of evidence for interventions included in this review (see table ). The categorisation of the quality of the evidence (high, moderate, low, or very low) reflects the quality of evidence available for our chosen outcomes in our defined populations of interest. These categorisations are not necessarily a reflection of the overall methodological quality of any individual study, because the Clinical Evidence population and outcome of choice may represent only a small subset of the total outcomes reported, and population included, in any individual trial. For further details of how we perform the GRADE evaluation and the scoring system we use, please see our website (www.clinicalevidence.com).

Table.

GRADE evaluation of interventions for Obesity in adults

Important outcomes Weight loss, mortality, adverse effects
Number of studies (participants) Outcome Comparison Type of evidence Quality Consistency Directness Effect size GRADE Comment
What are the effects of drug treatments in adults with obesity?
At least 28 (at least 11,087) Weight loss Orlistat v placebo 4 –1 0 0 0 Moderate Quality point deducted for incomplete reporting of results in one systematic review (number of people in analysis not reported)
At least 26 (at least 10,095) Adverse effects Orlistat v placebo 4 –1 0 0 0 Moderate Quality point deducted for incomplete reporting of results
6 (368) Weight loss Phentermine v placebo 4 –1 0 0 0 Moderate Quality point deducted for incomplete reporting of results
1 (99) Weight loss Phentermine v diethylpropion 4 –2 0 0 0 Low Quality points deducted for sparse data and incomplete reporting of results
4 (at least 4099) Weight loss Rimonabant v placebo 4 0 0 0 0 High
3 (2590) Weight loss Different doses of rimonabant v each other 4 0 0 –1 0 Moderate Directness point deducted for low number of comparators
4 (at least 4155) Adverse effects Rimonabant v placebo 4 0 0 0 0 High
4 (5023) Adverse effects Different doses of rimonabant v each other 4 –1 0 –1 0 Low Quality point deducted for incomplete reporting of results. Directness point deducted for low number of comparators
At least 16 (at least 5508) Weight loss Sibutramine v placebo 4 –1 0 0 0 Moderate Quality point deducted for methodological issues (for inclusion of unpublished studies in one systematic review and combined analysis of weight loss and maintenance in one systematic review)
5 (795) Weight loss Sibutramine v orlistat 4 –1 –1 0 0 Low Quality point deducted for methodological issues (heterogeneity among RCTs and short follow-up in some included RCTs [less than 4 months]). Consistency point deducted for conflicting results for different populations
1 (89) Weight loss Sibutramine plus orlistat v orlistat 4 –1 0 0 0 Moderate Quality point deducted for sparse data
2 (123) Weight loss Sibutramine plus orlistat v sibutramine 4 –3 0 0 0 Very low Quality points deducted for sparse data, incomplete reporting of results, and poor follow-up
What are the effects of bariatric surgery in adults with morbid obesity?
5 (3757) Weight loss Bariatric surgery v non-surgical treatments 4 –1 0 –1 0 Low Quality point deducted for inclusion of observational data. Directness point deducted for restricted population (predominantly white women) that may affect generalisability of results
5 (351) Weight loss Gastric bypass v vertical banded gastroplasty 4 –1 0 0 0 Moderate Quality point deducted for incomplete reporting of results
1 (51) Weight loss Gastric bypass v gastric banding 4 –1 0 0 0 Moderate Quality point deducted for sparse data
1 (32) Weight loss Gastric bypass v sleeve gastrectomy 4 –1 0 0 0 Moderate Quality point deducted for sparse data
3 (296) Weight loss Proximal v distal gastric bypass 4 –1 0 0 0 Moderate Quality point deducted for incomplete reporting of results
3 (309) Weight loss Open v laparoscopic gastric bypass 4 –2 0 0 0 Low Quality points deducted for incomplete reporting of results and for not carrying out a between group statistical assessment
3 (259) Weight loss Gastric banding v vertical banded gastroplasty 4 –2 –1 0 0 Very low Quality points deducted for incomplete reporting and not carrying out a between-group statistical assessment. Consistency point deducted for different results for different outcomes and at different time frames
1 (80) Weight loss Gastric banding v sleeve gastrectomy 4 –1 0 –1 0 Low Quality point deducted for sparse data. Directness point deducted for inclusion of people who were not morbidly obese, which may affect generalisability of results
1 (50) Weight loss Open v laparoscopic gastric banding 4 –2 0 0 0 Low Quality points deducted for sparse data and incomplete reporting of results
1 (30) Weight loss Open v laparoscopic vertical banded gastroplasty 4 –2 0 0 0 Low Quality points deducted for sparse data and for not carrying out a statistical assessment
Open in a new tab

Type of evidence: 4 = RCT; 2 = Observational. Consistency: similarity of results across studies.Directness: generalisability of population or outcomes.Effect size: based on relative risk or odds ratio.

Glossary

Bariatrics

The branch of medicine concerned with the management (prevention and control) of obesity and its related diseases.

Biliopancreatic diversion

There are two different types of biliopancreatic diversion. Standard biliopancreatic diversion surgically removes the lower third of the stomach and then forms a connection with the remaining stomach pouch with a portion of the small intestine beyond where the stomach was originally attached. Biliopancreatic diversion with duodenal switch divides the stomach vertically and removes the left half, leaving the connection between the stomach and the duodenum of the small intestine intact. A length of intestine is also removed and the duodenum is reconnected further down the small intestine. The aim is to increase weight loss by reducing calories and decreasing nutrient absorption.

Body mass index (BMI)

Expressed as weight in kilograms divided by height in metres squared (kg/m2). In the USA and UK, individuals with BMIs of 25–30 kg/m2 are considered overweight; those with BMIs above 30 kg/m2 are considered obese.

Gastric bypass

The roux-en-Y gastric bypass procedure involves dividing the stomach and creating a small pouch, which is then closed using several rows of staples. The remaining portion of the stomach is not removed but is “bypassed” and plays a diminished role in the digestive process. A Y-shaped portion of the small intestine is then attached to the pouch. The volume the new stomach pouch is capable of holding is about 25 g. The aim is to increase weight loss by reducing calories, altering GI appetite hormones, and decreasing nutrient absorption.

Gastroplasty

Vertical banded gastroplasty involves stapling the front of the stomach to the back of the stomach along a vertical plane, partitioning the stomach into two unequal parts that connect through a small (about 0.5 cm) opening. This allows the partially digested food to move from the small stomach pouch into the rest of the stomach and then the intestines. The newly created upper pouch will only allow the person to consume small amounts of food at a time.

High-quality evidence

Further research is very unlikely to change our confidence in the estimate of effect.

Low-quality evidence

Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.

Moderate-quality evidence

Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.

Very low-quality evidence

Any estimate of effect is very uncertain.

Obesity in children.

Effects of weight loss in preventing cardiovascular disease in the general population: see review on primary prevention of CVD: diet and weight loss.

Disclaimer

The information contained in this publication is intended for medical professionals. Categories presented in Clinical Evidence indicate a judgement about the strength of the evidence available to our contributors prior to publication and the relevant importance of benefit and harms. We rely on our contributors to confirm the accuracy of the information presented and to adhere to describe accepted practices. Readers should be aware that professionals in the field may have different opinions. Because of this and regular advances in medical research we strongly recommend that readers' independently verify specified treatments and drugs including manufacturers' guidance. Also, the categories do not indicate whether a particular treatment is generally appropriate or whether it is suitable for a particular individual. Ultimately it is the readers' responsibility to make their own professional judgements, so to appropriately advise and treat their patients.To the fullest extent permitted by law, BMJ Publishing Group Limited and its editors are not responsible for any losses, injury or damage caused to any person or property (including under contract, by negligence, products liability or otherwise) whether they be direct or indirect, special, incidental or consequential, resulting from the application of the information in this publication.

Contributor Information

David DeLaet, Mount Sinai School of Medicine, New York, USA.

Daniel Schauer, Institute for the Study of Health, Cincinnati, USA.

References

  • 1.National Institutes of Health. Clinical guidelines on the identification, evaluation, and treatment of overweight and obesity in adults: the Evidence Report. Bethesda, MD: US Department of Health and Human Services, 1998. [Google Scholar]
  • 2.World Health Organization. Obesity: preventing and managing the global epidemic. Report of a WHO consultation. WHO Technical Series. Geneva: World Health Organization, 2000; No. 894. [PubMed] [Google Scholar]
  • 3.Khan LK, Serdula MK, Bowman BA, et al. Use of prescription weight loss pills among US adults in 1996–1998. Ann Intern Med 2001;134:282–286. [DOI] [PubMed] [Google Scholar]
  • 4.Rennie KL, Jebb SA. Prevalence of obesity in Great Britain. Obes Rev 2005;6:11–12. [DOI] [PubMed] [Google Scholar]
  • 5.Flegal KM, Carroll MD, Ogden CL, et al. Prevalence and trends in obesity among US adults, 1999–2000. JAMA 2002;288:1723–1727. [DOI] [PubMed] [Google Scholar]
  • 6.Ogden CL, Carroll MD, McDowell MA, et al. Obesity among adults in the United States - no statisically significant change since 2003-2004. NCHS data brief no. 1. Hyattsville, MD: National Center for Health Statistics, 2007. Available at www.cdc.gov/nchs/data/databriefs/db01.pdf (last accessed 6 January 2010). [PubMed] [Google Scholar]
  • 7.Schwartz MW, Woods SC, Porte D, et al. Central nervous system control of food intake. Nature 2000;404:661–671. [DOI] [PubMed] [Google Scholar]
  • 8.Weinsier RL, Hunter GR, Heini AF, et al. The etiology of obesity: relative contribution of metabolic factors, diet, and physical activity. Am J Med 1998;105:145–150. [DOI] [PubMed] [Google Scholar]
  • 9.French SA, Story M, Jeffery RW. Environmental influences on eating and physical activity. Annu Rev Public Health 2001;22:309–335. [DOI] [PubMed] [Google Scholar]
  • 10.Pinkerton JH, Kopelman PG. Endocrine determinants of obesity. In: Bray GA, Bouchard C, eds. Handbook of obesity: etiology and pathophysiology. 2nd ed. New York, NY: Marcel Dekker, 2004: 655–669. [Google Scholar]
  • 11.Quesenberry CP, Caan B, Jacobson A. Obesity, health services use, and health care costs among members of a health maintenance organization. Arch Intern Med 1998;158:466–472. [DOI] [PubMed] [Google Scholar]
  • 12.Kushner RF, Foster GD. Obesity and quality of life. Nutrition 2000;16:947–952. [DOI] [PubMed] [Google Scholar]
  • 13.Vasan RS, Pencina MJ, Cobain M, et al. Estimated risks for developing obesity in the Framingham Heart Study. Ann Intern Med 2005;143:473–480. [DOI] [PubMed] [Google Scholar]
  • 14.Li Z, Maglione M, Tu W, et al. Meta-analysis: pharmacologic treatment of obesity. Ann Intern Med 2005;142:532–546. Search date 2003. [DOI] [PubMed] [Google Scholar]
  • 15.Hutton B, Fergusson D. Changes in body weight and serum lipid profile in obese patients treated with orlistat in addition to a hypocaloric diet: a systematic review of randomized clinical trials. Am J Clin Nutr 2004;80:1461–1468. Search date 2004. [DOI] [PubMed] [Google Scholar]
  • 16.Padwal R, Rucker D, Li S, et al. Long-term pharmacotherapy for obesity and overweight. In: The Cochrane Library, Issue 2, 2009. Chichester, UK: John Wiley & Sons, Ltd. Search date 2006. [Google Scholar]
  • 17.Cocco G, Pandolfi S, Rousson V. Sufficient weight reduction decreases cardiovascular complications in diabetic patients with the metabolic syndrome: a randomized study of orlistat as an adjunct to lifestyle changes (diet and exercise). Heart Drug 2005;5:68–74. [Google Scholar]
  • 18.Erdman J, Lippl F, Klose G, et al. Cholesterol lowering effect of dietary weight loss and orlistat treatment – efficacy and limitations. Aliment Pharmacol Ther 2004;19:1173–1179. [DOI] [PubMed] [Google Scholar]
  • 19.Guy-Grand B, Drouin P, Eschwege E, et al. Effects of orlistat on obesity-related diseases – a six-month randomized trial. Diabetes Obes Metab 2004;6:375–383. [DOI] [PubMed] [Google Scholar]
  • 20.Golay A, Laurent-Jaccard A, Habicht F, et al. Effect of orlistat in obese patients with binge eating disorder. Obes Res 2005;13:1701–1708. [DOI] [PubMed] [Google Scholar]
  • 21.Joffe G, Takala P, Tchoukhine E, et al. Orlistat in clozapine- or olanzapine-treated patients with overweight or obesity: a 16-week randomized, double-blind, placebo-controlled trial. J Clin Psychiatry 2008;69:706–711. [DOI] [PubMed] [Google Scholar]
  • 22.Haddock CK, Poston WSC, Dill PL, et al. Pharmacotherapy for obesity: a quantitative analysis of four decades of published randomized clinical trials. Int J Obes 2002;26:262–273. Search date 1999. [DOI] [PubMed] [Google Scholar]
  • 23.Wadden TA, Berkowitz RI, Womble LG, et al. Effects of sibutramine plus orlistat in obese women following 1 year of treatment by sibutramine alone: a placebo-controlled trial. Obes Res 2000;8:431–437. [DOI] [PubMed] [Google Scholar]
  • 24.Neovius M, Johansson K, Rossner S, et al. Head-to-head studies evaluating efficacy of pharmaco-therapy for obesity: a systematic review and meta-analysis. Obes Rev 2008;9:420–427. [DOI] [PubMed] [Google Scholar]
  • 25.Gaasch WH, Aurigemma GP. Valvular heart disease induced by anorectic drugs. In: UpToDate, Issue 8/3. Wellesley, MA: UpToDate Inc., 2003. [Google Scholar]
  • 26.Medicines Control Agency. Committee on Safety in Medicines. Important safety message: European withdrawal of anorectic agents/appetite suppressants: new legal developments, no new safety issues: licences for phentermine and amfepramone being withdrawn May 2001. Information page. Available at http://www.mhra.gov.uk/Safetyinformation/Safetywarningsalertsandrecalls/Safetywarningsandmessagesformedicines/CON019540 (last accessed 5 January 2010). [Google Scholar]
  • 27.Curioni C and André C. Rimonabant for overweight or obesity. In: The Cochrane Library, Issue 2, 2009. Chichester, UK: John Wiley & Sons, Ltd. Search date 2009. 17054276 [Google Scholar]
  • 28.Nissen SE, Nicholls SJ, Wolski K, et al. Effect of rimonabant on progression of atherosclerosis in patients with abdominal obesity and coronary artery disease: the STRADIVARIUS randomized controlled trial. JAMA 2008;299:1547–1560. [DOI] [PubMed] [Google Scholar]
  • 29.European Medicines Agency. Withdrawal of the marketing authorisation in the European Union, 2009. http://www.emea.europa.eu/humandocs/PDFs/EPAR/zimulti/3956009en.pdf (last accessed 5 January 2010). [Google Scholar]
  • 30.Arterburn DE, Crane PK, Veenstra DL. The efficacy and safety of sibutramine for weight loss: a systematic review. Arch Intern Med 2004;164:994–1003. Search date 2002. [DOI] [PubMed] [Google Scholar]
  • 31.Hung YJ, Chen YC, Pei D, et al. Sibutramine improves insulin sensitivity without alteration of serum adiponectin in obese subjects with Type 2 diabetes. Diabet Med 2005;22:1024–1030. [DOI] [PubMed] [Google Scholar]
  • 32.Shechter M, Beigel R, Freimark D, et al. Short-term sibutramine therapy is associated with weight loss and improved endothelial function in obese patients with coronary artery disease. Am J Card 2006;97:1650–1653. [DOI] [PubMed] [Google Scholar]
  • 33.Derosa G, D'Angelo A, Salvadeo SA, et al. Sibutramine effect on metabolic control of obese patients with type 2 diabetes mellitus treated with pioglitazone. Metabolism 2008:1552–1557. [DOI] [PubMed] [Google Scholar]
  • 34.Scholze J, Grimm E, Herrmann D, et al. Optimal treatment of obesity-related hypertension: the Hypertension-Obesity-Sibutramine (HOS) study. Circulation 2007;115:1991–1998. [DOI] [PubMed] [Google Scholar]
  • 35.Health Sciences Authority. Centre for Pharmaceutical Administration. Drug Alerts. Updates report on sibutramine. Information page. Available at http://www.hsa.gov.sg/publish/hsaportal/en/health_products_regulation/safety_information/product_safety_alerts/safety_alerts_2002/sibutramine.html (last accessed 5 January 2010). [Google Scholar]
  • 36.Zannad F, Gille B, Grentzinger A, et al. Effects of sibutramine on ventricular dimensions and heart valves in obese patients during weight reduction. Am Heart J 2002;144:508–515. [DOI] [PubMed] [Google Scholar]
  • 37.Bach DS, Rissanen AM, Mendel CM, et al. Absence of cardiac valve dysfunction in obese patients treated with sibutramine. Obes Res 1999;7:363–369. [DOI] [PubMed] [Google Scholar]
  • 38.Sari R, Balci MK, Cakir M, et al. Comparison of efficacy of sibutramine or orlistat versus their combination in obese women. Endocr Res 2004;30:159–167. [DOI] [PubMed] [Google Scholar]
  • 39.Colquitt J, Picot J, Loveman E, et al. Surgery for morbid obesity. In: The Cochrane Library, Issue 2, 2009. Chichester, UK: John Wiley & Sons, Ltd. Search date 2008. [Google Scholar]
  • 40.Maggard MA, Shugarman LR, Suttorp M, et al. Meta-analysis: surgical treatment of obesity. Ann Intern Med 2005;142:547–559. Search date 2003. [DOI] [PubMed] [Google Scholar]
  • 41.Inabnet WB, Quinn T, Gagner M, et al. Laparoscopic Roux-en-Y gastric bypass in patients with BMI less than 50: a prospective randomized trial comparing short and long limb lengths. Obes Surg 2005;15:51–57. [DOI] [PubMed] [Google Scholar]
  • 42.Choban PS, Flancbaum L. The effect of Roux limb lengths on outcome after Roux-en-Y gastric bypass: a prospective, randomized clinical trial. Obes Surg 2002;12:540–545. [DOI] [PubMed] [Google Scholar]
  • 43.Pinheiro JS, Schiavon CA, Pereira PB, et al. Long-long limb Roux-en-Y gastric bypass is more efficacious in treatment of type 2 diabetes and lipid disorders in super-obese patients. Surg Obes Relat Dis 2008;4:521–525. [DOI] [PubMed] [Google Scholar]
  • 44.Lee WJ, Huang MT, Yu PJ, et al. Laparoscopic vertical banded gastroplasty and laparoscopic gastric bypass: a comparison. Obes Surg 2004;14:626–634. [DOI] [PubMed] [Google Scholar]
  • 45.Olbers T, Fagevik-Olsen M, Maleckas A, et al. Randomized clinical trial of laparoscopic Roux-en-Y gastric bypass versus laparoscopic vertical banded gastroplasty for obesity. Br J Surg 2005;92:557–562. [DOI] [PubMed] [Google Scholar]
  • 46.Nguyen NT, Longoria M, Welbourne S, et al. Glycolide copolymer staple-line reinforcement reduces staple site bleeding during laparoscopic gastric bypass: a prospective randomized trial. Arch Surg 2005;140:773–778. [DOI] [PubMed] [Google Scholar]
  • 47.Miller KA, Pump A. Use of bioabsorbable staple reinforcement material in gastric bypass: a prospective randomized clinical trial. Surg Obes Relat Dis 2007;3:417–421. [DOI] [PubMed] [Google Scholar]
  • 48.Karamanakos SN, Vagenas K, Kalfarentzos F, et al. Weight loss, appetite suppression, and changes in fasting and postprandial ghrelin and peptide-YY levels after Roux-en-Y gastric bypass and sleeve gastrectomy: a prospective, double blind study. Ann Surg 2008;247:401–407. [DOI] [PubMed] [Google Scholar]
  • 49.Angrisani L, Lorenzo M, Borrelli V. Laparoscopic adjustable gastric banding versus Roux-en-Y gastric bypass: 5-year results of a prospective randomized trial. Surg Obes Relat Dis 2007;3:127–132. [DOI] [PubMed] [Google Scholar]
  • 50.Davila-Cervantes A, Borunda D, Dominguez-Cherit G, et al. Open versus laparoscopic vertical banded gastroplasty: a randomized controlled double blind trial. Obes Surg 2002;12:812–818. [DOI] [PubMed] [Google Scholar]
  • 51.Himpens J, Dapri G, Cadiere GB. A prospective randomized study between laparoscopic gastric banding and laparoscopic isolated sleeve gastrectomy: Results after 1 and 3 years. Obes Surg 2006;16:1450–1456. [DOI] [PubMed] [Google Scholar]
  • 52.Dixon JB. Adjustable gastric banding and conventional therapy for type 2 diabetes: a randomized controlled trial. JAMA 2008;299:316–323. [DOI] [PubMed] [Google Scholar]
  • 53.Mingrone G, Greco AV, Giancaterini A, et al. Sex hormone-binding globulin levels and cardiovascular risk factors in morbidly obese subjects before and after weight reduction induced by diet or malabsorptive surgery. Atherosclerosis 2002;161:455–462. [DOI] [PubMed] [Google Scholar]
BMJ Clin Evid. 2010 Feb 11;2010:0604.

Orlistat

Summary

WEIGHT LOSS Compared with placebo: Orlistat is more effective at increasing weight loss at 6 to 12 months in obese people who are on a low-calorie diet irrespective of having diabetes, hyperlipidaemia, hypertension, or binge eating disorder, but its effects in people with weight gain related to treatment with antipsychotic drugs are less clear ( moderate-quality evidence ). Compared with sibutramine: Orlistat may be less effective at increasing weight loss in obese people without comorbidities, but we don't know how orlistat and sibutramine compare in people with type 2 diabetes or hypertension ( low-quality evidence ). Compared with orlistat plus sibutramine: Orlistat seems less effective at increasing weight loss at 6 months in obese people who are also on a reduced-calorie diet (moderate-quality evidence). ADVERSE EFFECTS Compared with placebo: Orlistat is associated with an increased risk of GI adverse effects such as diarrhoea, flatulence, bloating, and abdominal pain (moderate-quality evidence). NOTE We found no clinically important results from RCTs on orlistat compared with diethylpropion, mazindol, or phentermine in obese people.

Benefits

Orlistat versus placebo:

We found three systematic reviews (search dates 2003, 2004, and 2006), which between them identified 23 RCTs (11 of which were identified by all 3 reviews) comparing orlistat versus placebo. We also found four additional RCTs, and one subsequent RCT. In all the RCTs identified by the reviews, and in the four additional RCTs, people treated with orlistat also followed a low-calorie diet.

The reviews meta-analysed results from RCTs with similar study design and dose of orlistat. The reviews differed slightly in their inclusion criteria for minimum BMI (varied from 25 to 30 kg/m2 or more than 27 kg/m2 plus one or more obesity-related comorbidity), and one review specified a minimum follow-up of 1 year (not set in the other reviews). One review reported subgroup analysis in people with comorbidities. We have therefore reported results from all three reviews.

All three reviews found that orlistat significantly increased weight loss at 1 year compared with placebo, with mean weight loss being similar in all three reviews (mean weight loss ranged from 2.83 kg to 2.89 kg).

The first review found that orlistat significantly increased the mean difference in weight loss at 12 months compared with placebo in people with a minimum BMI of 27 kg/m2 (22 RCTs; number of people in analysis not reported; mean difference in weight loss: –2.89 kg, 95% CI –3.51 kg to –2.27 kg).

The second review also found that orlistat significantly increased mean weight loss at 12 months compared with placebo in people with a minimum BMI of greater than 25 kg/m2 (17 RCTs, 10,041 people; WMD in weight loss –2.83 kg, 95% CI –3.54 kg to –2.13 kg). The review also performed separate meta-analyses comparing orlistat versus placebo in people with type 2 diabetes mellitus and other high-risk cardiovascular risk factors (dyslipidaemia or hypertension). It found that orlistat significantly increased mean weight loss compared with placebo in people with type 2 diabetes mellitus, and in people with other high-risk cardiovascular factors (type 2 diabetes: 4 RCTs, 1729 people; WMD in weight loss –2.50 kg, 95% CI –2.97 kg to –2.02 kg; high-risk population: 4 RCTs, 1480 people; WMD in weight loss –2.04 kg, 95% CI –3.21 kg to –0.88 kg).

The third review found that orlistat significantly increased mean weight loss at 12 months compared with placebo in people with a minimum BMI of 30 kg/m2 or greater than 27 kg/m2 plus one or more obesity-related comorbidity (14 RCTs, 9457 people; WMD in weight loss –2.87 kg, 95% CI –3.21 kg to –2.53 kg). The review also found that a significantly larger proportion of people lost 5% or more of their initial body weight with orlistat compared with placebo (14 RCTs, 9389 people; 2652/4931 [54%] with orlistat v 1468/4458 [33%] with placebo; risk difference 0.21, 95% CI 0.18 to 0.24).

The first additional RCT (90 obese people with type 2 diabetes, hypertension, and metabolic syndrome) found that orlistat (120 mg 3 times daily) significantly increased weight loss at 6 months compared with placebo (mean weight loss: –5.6 kg with orlistat v –2.6 kg with placebo; P less than 0.001).

The second additional RCT (384 obese people with dyslipidaemia) found that orlistat (120 mg 3 times daily) significantly increased weight loss at 6 months compared with placebo (mean weight loss: –7.4 kg with orlistat v –4.9 kg with placebo; P less than 0.01).

The third additional RCT (1004 obese people with hypertension [614 people], dyslipidaemia [197 people], or type 2 diabetes [193 people]) found that orlistat (120 mg 3 times daily) significantly increased weight loss at 6 months compared with placebo in all populations (mean weight loss in people with type 2 diabetes: –3.9 kg with orlistat v –1.3 kg with placebo; hypertension: –5.8 kg with orlistat v –1.8 kg with placebo; dyslipidaemia: –5.0 kg with orlistat v –2.1 kg with placebo; P less than 0.0001 for all comparisons).

The fourth additional RCT (89 obese people with binge eating disorder) found that orlistat (120 mg 3 times daily) significantly increased weight loss at 24 weeks compared with placebo (mean % weight lost of initial body weight: 7% with orlistat v 2% with placebo; P = 0.0001).

The subsequent RCT (63 obese people having treatment with clozapine or olanzapine for a serious mental condition) found no significant difference between orlistat (120 mg 3 times daily) and placebo in weight loss at 16 weeks (mean change in weight: –1.25 kg with orlistat v +0.44 kg with placebo; P = 0.101). Unlike other RCTs of orlistat, people enrolled in the trial, although encouraged to limit their calorie intake, continued on their usual diet. Subgroup analysis based on gender found that, compared with placebo, orlistat significantly increased weight loss at 16 weeks in men (41 men; mean change in weight: –2.36 kg with orlistat v +0.62 kg with placebo; P = 0.011), but not in women (22 women; mean change in weight: +1.94 kg with orlistat v +0.22 kg with placebo; P = 0.397). The RCT also found no significant difference between groups in the proportion of people losing 5% or more of their initial body weight (5/31 [16%] with orlistat v 2/32 [6%] with placebo; P = 0.257).

Orlistat versus sibutramine:

See benefits of sibutramine.

Orlistat versus other drugs:

We found one systematic review (search date 1999), which identified no RCTs comparing orlistat versus diethylpropion, mazindol, or phentermine.

Orlistat alone versus orlistat plus sibutramine:

See benefits of sibutramine plus orlistat.

Harms

Orlistat versus placebo:

The first review found that orlistat was associated with a significantly higher risk of GI adverse effects including diarrhoea, flatulence, bloating, abdominal pain, and dyspepsia after 6 to 12 months of treatment compared with placebo (29 RCTs; diarrhoea: OR 54.85, 95% CI 44.88 to 67.48; flatulence: OR 3.72, 95% CI 3.16 to 4.39; bloating, abdominal pain, and dyspepsia: OR 1.55, 95% CI 1.18 to 2.06; absolute numbers not reported for any outcome).

The second review also found that orlistat was associated with a significantly higher risk of GI adverse effects at 1 year compared with placebo (16 RCTs, 9558 people; RR of at least one GI adverse effect: 1.46, 95% CI 1.37 to 1.55; absolute numbers not reported; details of GI adverse effects not described).

The third review found that orlistat was associated with a significantly higher rate of GI adverse effects compared with placebo (14 RCTs, 8938 people; 3805/4684 [81%] with orlistat v 2407/4254 [57%] with placebo; risk difference 0.24, 95% CI 0.20 to 0.29). The most commonly reported GI adverse effects reported were fatty/oily stool, faecal urgency, and oily spotting.

The first additional RCT found that orlistat increased the proportion of people with GI adverse events at 4 weeks (36% with orlistat v 24% with placebo; absolute numbers not reported; significance not assessed). The second additional RCT gave no information on adverse effects. The third additional RCT found that orlistat was associated with a significantly higher withdrawal rate because of adverse GI effects (mainly defecation difficulties) compared with placebo (10/499 [2.0%] with orlistat v 2/505 [0.4%] with placebo; P = 0.01). The fourth additional RCT gave no information on adverse effects.

The subsequent RCT found that a similar proportion in the orlistat and placebo groups experienced diarrhoea (11/31 [35%] with orlistat v 9/32 [28%] with placebo; significance not assessed).

Orlistat versus sibutramine:

See harms of sibutramine.

Orlistat versus other drugs:

We found no RCTs.

Orlistat alone versus orlistat plus sibutramine:

See harms of sibutramine plus orlistat.

Comment

Because of the high rates of GI adverse effects associated with orlistat, it remains unclear whether blinded evaluation of its effects is possible. At the end of a double-blinded 16-week trial, 22/26 (85%) people correctly identified their treatment group.

Substantive changes

Orlistat One systematic review added supports existing evidence on the effects of orlistat in adults who are obese. It found that orlistat increased mean weight loss at 12 months compared with placebo, but it was associated with an increased risk of GI adverse effects. One subsequent RCT added assessing the effects of orlistat in people with clozapine- or olanzapine-related weight gain found no significant difference between orlistat and weight loss at 16 weeks. However, subgroup analyses found that orlistat increased weight loss in men but not women. One systematic review added comparing orlistat versus sibutramine found that sibutramine increased weight loss compared with orlistat, but it found no significant difference between treatments in people with type 2 diabetes or hypertension. Categorisation unchanged (Trade-off between benefits and harms).

BMJ Clin Evid. 2010 Feb 11;2010:0604.

Phentermine

Summary

WEIGHT LOSS Compared with placebo: Phentermine seems more effective at increasing weight loss in obese people undertaking additional lifestyle interventions ( moderate-quality evidence ). Compared with diethylpropion: Phentermine may be more effective at increasing weight loss ( low-quality evidence ). ADVERSE EFFECTS Phentermine may be associated with heart and lung problems. NOTE We found no direct information from RCTs about weight regain and long-term safety of phentermine in the treatment of adults with obesity. We found no clinically important results from RCTs about phentermine compared with orlistat or sibutramine in obese people.

Benefits

Phentermine versus placebo:

We found one systematic review (search date 1999; 6 RCTs, 368 people) comparing phentermine 15 mg to 30 mg daily versus placebo in obese adults, with a mean follow-up of 13.2 weeks (range 2–24 weeks). Most of the people treated with phentermine received additional lifestyle interventions. The review found that phentermine produced significantly more weight loss compared with placebo (effect size less than 0.6 [information presented graphically]; mean difference in weight loss between phentermine and placebo in the 6 RCTs ranged from 0.6 kg to 6.0 kg).

Phentermine versus diethylpropion:

We found one systematic review that found that phentermine significantly increased weight loss compared with diethylpropion (search date 1999; 1 RCT, 99 people; mean weight loss: –8.3 kg with phentermine v –6.3 kg with diethylpropion; effect size: 0.57, CI not reported).

Phentermine versus other drugs:

We found one systematic review, which found no RCTs comparing phentermine versus orlistat, rimonabant, or sibutramine.

Harms

Phentermine versus placebo/diethylpropion/other drugs:

The systematic review gave no information on adverse effects.

Phentermine given alone has not been associated with valvular heart disease. A European Commission review reported that, although no new safety problems were identified with phentermine, a link between phentermine and heart and lung problems could not be totally excluded.

Comment

Data regarding the safety and efficacy of phentermine are limited compared with orlistat and sibutramine, and high withdrawal rates have been reported for phentermine. These factors should be considered when making decisions regarding the use of phentermine.

Substantive changes

No new evidence

BMJ Clin Evid. 2010 Feb 11;2010:0604.

Rimonabant

Summary

WEIGHT LOSS Compared with placebo: Rimonabant (at doses of 20 mg and 5 mg) is more effective at increasing weight loss at 1 year in people who are obese and on a mild hypocaloric diet ( high-quality evidence ). Different doses of rimonabant compared with each other: Rimonabant 20 mg seems more effective than rimonabant 5 mg at increasing weight loss at 1 year in people who are obese and on a mild hypocaloric diet ( moderate-quality evidence ). ADVERSE EFFECTS Compared with placebo: Rimonabant 20 mg is associated with increased risk of adverse effects such as depressed mood, anxiety, headache, dizziness, paraesthesia, nausea, and vomiting, but rimonabant 5 mg seems to be associated with a lower risk of adverse effects, with rates similar to those with placebo (high-quality evidence). Different doses of rimonabant compared with each other: Rimonabant 20 mg may be associated with a higher rate of discontinuation because of adverse effects compared with rimonabant 5 mg, but may be associated with a similar risk of serious adverse effects (not further defined) compared with rimonabant 5 mg ( low-quality evidence ).

Benefits

Rimonabant versus placebo:

We found two systematic reviews (both with search date 2006) and one subsequent RCT. Both reviews identified the same four RCTs but meta-analysed data for different outcomes and comparisons, so we report both reviews here. All four RCTs (6635 people with BMI more than 27 kg/m2, some of whom had type 2 diabetes mellitus, hypertension, or dyslipidaemia) identified by the reviews compared rimonabant 20 mg or rimonabant 5 mg daily versus placebo; all people enrolled in the RCTs were also advised to undertake a mild hypocaloric diet (600 kcal/day deficit).

The reviews found that rimonabant 20 mg significantly increased weight loss at 12 months compared with placebo (meta-analysis from first review; 4 RCTs, 4099 people; WMD –4.67 kg, 95% CI –5.26 kg to –4.07 kg). The first review also reported that a significantly larger proportion of people achieved weight loss of 5% or more of their initial body weight with rimonabant 20 mg daily compared with placebo (4 RCTs, 4099 people; 1265/2500 [51%] with rimonabant v 297/1599 [19%] with placebo; risk difference 0.33, 95% CI 0.29 to 0.37).

The second review found that rimonabant 5 mg daily significantly increased weight loss at 1 year compared with placebo (3 RCTs, 2301 people; WMD –1.25 kg, 95% CI –1.64 kg to –0.86 kg).

Different doses of rimonabant versus each other:

The second review found that rimonabant 20 mg daily significantly increased weight loss at 1 year compared with rimonabant 5 mg daily (3 RCTs, 2590 people; WMD –3.29 kg, 95% CI –3.72 kg to –2.86 kg).

Rimonabant versus other drugs:

We found no RCTs.

Harms

In January 2009, the European Medicines Agency (EMEA) suspended marketing authorisation of rimonabant in the European Union because of the increased risk of psychiatric disorders.

Rimonabant versus placebo:

The reviews found that rate of treatment discontinuation because of adverse effects was significantly higher with rimonabant 20 mg compared with placebo (meta-analysis from second review; 4 RCTs, 4105 people; 346/2503 [14%] with rimonabant v 115/1602 [8%] with placebo; RR 1.92, 95% CI 1.57 to 2.34). Most notably, the first review noted that rimonabant was associated with a significant increase in incidence of psychiatric disorders (depression, anxiety, irritability, and aggression) compared with placebo (4 RCTs, 4105 people; 148/2503 [6%] with rimonabant v 39/1602 [2%] with placebo; risk difference 0.03, 95% CI 0.02 to 0.05).

The second review found no significant difference in adverse effects between rimonabant 5 mg and placebo (4 RCTs, 4122 people; 137/2520 [5%] with rimonabant v 69/1602 [4%] with placebo; OR 1.29, 95% CI 0.96 to 1.73).

Different doses of rimonabant versus each other:

The review found that rate of discontinuation because of adverse effects was significantly higher with rimonabant 20 mg compared with rimonabant 5 mg (4 RCTs, 5023 people; 346/2503 [14%] with rimonabant 20 mg v 221/2520 [9%] with rimonabant 5 mg; RR 1.58, 95% CI 1.34 to 1.85). However, there was no significant difference between groups in the proportion of people experiencing serious adverse effects (not further defined) (4 RCTs, 5023 people; 149/2503 [6%] with rimonabant 20 mg v 137/2520 [5%] with rimonabant 5 mg; RR 1.10, 95% CI 0.88 to 1.38).

Rimonabant versus other drugs:

We found no RCTs.

Comment

We found one large RCT (839 people requiring coronary angiography for a clinical indication who also had a waist circumference over 88 cm for women or over 102 cm for men, and either met pre-specified criteria for the presence of metabolic syndrome or were current smokers) assessing the effects of rimonabant on progression of atherosclerosis. The RCT did not meet Clinical Evidence reporting criteria (it did not assess weight loss as a primary or secondary outcome), but as this is a key RCT we have decided to report data on our outcome of interest and adverse effects here. The RCT found that rimonabant 20 mg significantly increased weight loss at 18 months compared with placebo (mean weight loss: –4.3 kg with rimonabant v –0.5 kg with placebo; P less than 0.001). However, the RCT found that rimonabant 20 mg was associated with a significantly higher rate of adverse psychiatric disorders (primarily an increase in anxiety and depression) compared with placebo (183/422 [43%] with rimonabant v 118/416 [28%] with placebo; P less than 0.001).

Substantive changes

Rimonabant One systematic review added augments evidence already reported that rimonabant increases weight loss at 12 months compared with placebo. The review also found that a larger proportion of people achieved weight loss of 5% or more of their initial body weight with rimonabant compared with placebo. However, more people experienced serious adverse effects with rimonabant compared with placebo, including psychiatric disorders (depression, anxiety, and irritability). In January 2009, the European Medicines Agency (EMEA) suspended marketing authorisation of rimonabant in the European Union because of the increased risk of psychiatric disorders. Categorisation unchanged (Trade-off between benefits and harms).

BMJ Clin Evid. 2010 Feb 11;2010:0604.

Sibutramine

Summary

WEIGHT LOSS Compared with placebo: Sibutramine is more effective at 4 to 12 months at increasing weight loss in obese people who are undertaking dietary interventions with or without exercise, irrespective of co-existing conditions such as diabetes, hypertension, hyperlipidaemia, binge eating disorder, or CAD ( moderate-quality evidence ). Compared with orlistat: Sibutramine may be more effective at increasing weight loss in obese people without comorbidities, but we don't know how sibutramine and orlistat compare in people with type 2 diabetes or hypertension ( low-quality evidence ). Compared with sibutramine plus orlistat: We don't know whether sibutramine alone is more effective at increasing weight loss at 6 months in obese people who are also on a reduced-calorie diet or at maintaining weight loss at 16 weeks in obese people with a modified lifestyle who have taken it for 1 year ( very low-quality evidence ). ADVERSE EFFECTS Sibutramine has been associated with increases in systolic and diastolic blood pressure, heart rate, and total as well as LDL cholesterol; conversely, it has been associated with modest decreases in triglyceride levels, fasting serum glucose levels, and glycosylated haemoglobin levels, and with modest increases in HDL cholesterol levels. In January 2010, the European Medicines Agency (EMEA) suspended marketing authorisation of sibutramine in the European Union because of the increased risk of non-fatal myocardial infarctions and strokes.

Benefits

Sibutramine versus placebo:

We found two systematic reviews, two additional RCTs, and two subsequent RCTs comparing sibutramine versus placebo. The two reviews, which between them identified 34 RCTs, differed in their inclusion criteria, and so include different RCTs in their meta-analysis. We report both reviews here. Most of the people treated with sibutramine received additional dietary interventions, and many also received an exercise intervention. The first review suggested that weight loss with sibutramine is associated with both positive and negative changes in cardiovascular and metabolic risk factors.

The first review (search date 2002; 29 RCTs [number of people not reported] in people with BMI 25–40 kg/m2, some with diabetes, hypertension, hyperlipidaemia, or binge eating disorder) included unpublished RCTs (authors of the review contacted key researchers in the field and pharmaceutical companies), and RCTs with a minimum follow-up of 8 weeks. The review meta-analysed data in three groups based on follow-up of 8 to 12 weeks, 16 to 24 weeks, and 44 to 54 weeks. For drug interventions, our minimum length of follow-up for reporting is 4 months, and so we report only meta-analysis of data at 16 weeks or more of follow-up. All the meta-analyses found that sibutramine significantly increased weight loss compared with placebo. Trials of 16 to 24 weeks' duration, all comparing sibutramine 10 mg to 15 mg daily versus placebo, were meta-analysed in three subgroups because of significant heterogeneity among the trials in methods of analysis (statistically heterogeneous; P less than 0.001). All the meta-analyses found that sibutramine significantly increased weight loss compared with placebo (4 RCTs [484 people] that used last-observation-carried-forward analysis and had more than 70% follow-up: WMD –3.43 kg, 95% CI –4.50 kg to –2.36 kg; 5 RCTs [473 people] that assessed completers: –6.03 kg, 95% CI –7.36 kg to –4.70 kg; 3 RCTs [222 people] with follow-up of less than 70%: –5.06 kg, 95% CI –6.16 kg to –3.96 kg); people who completed the trial had the greatest weight loss. The review also found that sibutramine 10 mg to 15 mg daily significantly increased weight loss at 45 to 54 weeks compared with placebo (5 RCTs, 2188 people; WMD –4.45 kg, 95% CI –5.29 kg to –3.62 kg). The review found similar rates of weight loss in trials that specifically recruited obese adults with type 2 diabetes mellitus, hypertension, or hyperlipidaemia, and in trials in obese adults who did not have comorbidities (data not reported in review).

The second review (search date 2006; 10 RCTs, 2348 people with BMI 30 kg/m2 or more than 27 kg/m2 plus one or more obesity-related comorbidity) included only published RCTs with a minimum length of follow-up of 1 year. The review found that sibutramine significantly increased mean weight loss at 12 months compared with placebo (10 RCTs, 2348 people; WMD weight loss: –4.16 kg, 95% CI –4.73 kg to –3.59 kg). The review also found that a significantly larger proportion of people lost 5% or more of their initial body weight with sibutramine compared with placebo (7 RCTs, 1464 people; 443/808 [57%] with sibutramine v 174/656 [25%] with placebo; risk difference 0.32, 95% CI 0.27 to 0.37). The review combined data on weight loss from RCTs separately assessing weight loss and weight maintenance in the same analysis.

The first additional RCT (48 obese adults with type 2 diabetes) compared sibutramine 15 mg daily versus placebo for 6 months. It found that sibutramine significantly increased mean weight loss at 6 months compared with placebo (–2.4 kg with sibutramine v –0.7 kg with placebo; P less than 0.05). The second additional RCT (80 obese adults with CAD) compared sibutramine 10 mg daily versus placebo for 4 months. It found that sibutramine significantly increased percentage weight reduction from initial body weight at 4 months compared with placebo (11% with sibutramine v 2% with placebo; P less than 0.001).

The first subsequent RCT (138 people with BMI 30 kg/m2 or more and who had type 2 diabetes mellitus and inadequate glycaemic control on a sulfonylurea or metformin) compared sibutramine 10 mg daily versus placebo for 6 months. It found that sibutramine significantly increased weight loss at 6 months compared with placebo (change in body weight from baseline: from 102 kg to 96 kg with sibutramine v from 101 kg to 100 kg with placebo; P less than 0.05). Before randomisation to sibutramine or placebo, all people enrolled completed a 3-month run-in period in which they took a fixed dose of pioglitazone, and started a low-calorie diet (about 600 kcal daily deficit). Treatment with pioglitazone continued for the duration of the study. The RCT reported no large changes in body weight in the run-in phase in either group. The second subsequent RCT (171 people with BMI of 27–45 kg/m2 and hypertension) compared sibutramine 15 mg daily versus placebo over 16 weeks. It found that sibutramine significantly increased mean weight loss at 16 weeks compared with placebo (–5.7 kg with sibutramine v –1.5 kg with placebo; P less than 0.0001). The RCT also found that a significantly larger proportion of people lost 5% or more of their initial body weight with sibutramine compared with placebo (54.7% with sibutramine v 14.5% with placebo; absolute numbers not reported; P less than 0.0001). General dietary and exercise advice was given to people enrolled in the RCT.

Sibutramine versus orlistat:

We found one systematic review (search date 2007; 8 RCTs, 885 people, some with diabetes or hypertension) comparing sibutramine versus orlistat. The review found that sibutramine (10 mg daily in 6 RCTs, 15 mg daily in 1 RCT, and 20 mg daily in 1 RCT) significantly increased weight loss compared with orlistat (120 mg 3 times daily) at 3 to 12 months (median duration 7 months) (7 RCTs, 795 people; WMD in weight loss –2.2 kg, 95% CI –3.9 kg to –0.5 kg). However, the review reported significant heterogeneity among the RCTs (statistically heterogeneous: P less than 0.001). Two of the RCTs included in the meta-analysis were subgroup analyses of people with type 2 diabetes or hypertension. When meta-analysing results in people with and without comorbidities separately, results of both meta-analyses were statistically homogeneous. Analysis of those RCTs not specifically assessing people with type 2 diabetes or hypertension found the difference in weight loss at 3 to 8.8 months between sibutramine and orlistat to be significant in favour of sibutramine (5 RCTs, 478 people; WMD in weight loss –3.4 kg, 95% CI –4.6 kg to –2.3 kg). However, in people with type 2 diabetes or hypertension, the review found no significant difference in weight loss at 12 months between sibutramine and orlistat (2 RCTs, 253 people; WMD in weight loss: +0.4 kg, 95% CI –0.6 kg to +1.4 kg).

Sibutramine versus diethylpropion, mazindol, rimonabant, or phentermine:

We found one systematic review (search date 1999), which identified no RCTs comparing sibutramine versus diethylpropion, mazindol, or phentermine.

Sibutramine plus orlistat:

See benefits of sibutramine plus orlistat.

Harms

Sibutramine was temporarily suspended from the market in Italy in March 2002 in response to 50 reported adverse reactions, including seven severe adverse reactions (tachycardia, hypertension, and arrhythmia) and two deaths resulting from cardiac arrest. The Central European Committee for Proprietary Medicinal Products completed a review of sibutramine in June 2002, and concluded that the risk–benefit profile of sibutramine remained in favour of benefit; it therefore lifted the suspension in August 2002.However, since the search date of this review, sibutramine has been evaluated by the European Medicines Agency. The Agency's Committee for Medicinal Products for Human Use (CHMP) concluded that the risks are greater than the benefits and recommended suspending marketing of sibutramine in the European Union in January 2010 (see drug safety alert below for further information).

Sibutramine versus placebo:

We found no evidence about adverse effects after more than 1 year of treatment.

General adverse effects:

The first review found that sibutramine was associated with increased total and LDL cholesterol levels at 16 to 24 weeks compared with placebo (increase in total cholesterol [data reported are mean difference of sibutramine minus placebo]: –1.9 mg/dL to +1.8 mg/dL; increase in LDL cholesterol: 0.6 mg/dL to 2.6 mg/dL); however, no difference from baseline levels was noted at 44 to 54 weeks.

The second review reported that insomnia, nausea, dry mouth, and constipation were more common in people receiving sibutramine, occurring at frequency rates of 7% to 20% (absolute numbers not reported; significance not assessed).

The first additional RCT found no significant difference between sibutramine and placebo in serum triglycerides, total cholesterol, LDL cholesterol, or HDL cholesterol (triglycerides: 1.9 mmol/L with sibutramine v 2.6 mmol/L with placebo; total cholesterol: 4.9 mmol/L in each group; LDL cholesterol: 3.0 mmol/L with sibutramine v 2.8 mmol/L with placebo; HDL cholesterol: 1.1 mmol/L with sibutramine v 1.0 mmol/L with placebo; all comparisons reported as not significant; P value not reported for any outcome).

Cardiovascular adverse effects:

The first review found that sibutramine increased systolic blood pressure and diastolic blood pressure compared with placebo (mean difference in systolic blood pressure [data reported are mean difference of sibutramine minus placebo]: range from –1.6 mm Hg to +5.6 mm Hg at 16–24 weeks in several RCTs, +4.6 mm Hg at 44–54 weeks; mean difference in diastolic blood pressure: +1.6 mm Hg at 8–12 weeks, –0.8 mm Hg to +1.7 mm Hg at 16–24 weeks, and +2.8 mm Hg at 44–54 weeks in several RCTs). The review found significant heterogeneity (statistically heterogeneous; P less than 0.001) among RCTs assessing blood pressure and reported change in systolic and diastolic blood pressures across RCTs. It also found that sibutramine significantly increased heart rate compared with placebo (increase in heart rate: 0.75–5.9 bpm at 16–24 weeks, and 5.9 bpm at 44–54 weeks).

The second review also found that, compared with placebo, sibutramine was associated with significant increases in systolic (7 RCTs, 1906 people; WMD 1.69 mm Hg, 95% CI 0.11 mm Hg to 3.28 mm Hg) and diastolic blood pressure at 12 months (7 RCTs, 1906 people; WMD 2.42 mm Hg, 95% CI 1.51 mm Hg to 3.32 mm Hg). Sibutramine was also associated with significant increase in heart rate at 12 months compared with placebo (7 RCTs, 1658 people; WMD in heart rate: 4.53 bpm, 95% CI 3.49 bpm to 5.57 bpm).

The first additional RCT found no significant difference between sibutramine and placebo in heart rate, systolic blood pressure, or diastolic blood pressure (heart rate: 74 bpm in both groups; systolic blood pressure: 135 mm Hg with sibutramine v 133 mm Hg with placebo; diastolic blood pressure: 87 mm Hg with sibutramine v 88 mm Hg with placebo; all comparisons reported as not significant; P value not reported for any outcome). The second additional RCT reported no differences in mean systolic or diastolic blood pressure, or in heart rate during the 4-month follow-up between the sibutramine and placebo groups (absolute numbers not reported; significance not assessed).

The first subsequent RCT found that the change in systolic blood pressure and diastolic blood pressure at 6 months was similar with sibutramine and placebo (change from baseline; systolic blood pressure: from 135.6 mm Hg to 133.2 mm Hg with sibutramine v from 136.1 mm Hg to 134.4 mm Hg with placebo; diastolic blood pressure: from 86.9 mm Hg to 84.0 mm Hg with sibutramine v from 85.9 mm Hg to 84.2 mm Hg with placebo; significance not assessed for either outcome). The RCT gave no information on other adverse effects. The second subsequent RCT found no significant difference between sibutramine and placebo in change in systolic or diastolic blood pressure at 16 weeks (mean change from baseline; systolic blood pressure: –5.9 mm Hg with sibutramine v –4.8 mm Hg with placebo; P = 0.5494; diastolic blood pressure: –0.7 mm Hg with sibutramine v –2.3 mm Hg with placebo; P = 0.1785). However, the RCT found that sibutramine was associated with a significant increase in 24-hour diastolic blood pressure at 16 weeks compared with placebo (mean change from baseline: +2.1 mm Hg with sibutramine v –0.3 mm Hg with placebo; P = 0.0403), but there was no significant difference between groups in 24-hour systolic blood pressure (–0.3 mm Hg with sibutramine v –0.9 mm Hg with placebo; P = 0.6939). Similarly, the RCT found that sibutramine was associated with a significant increase in heart rate at 16 weeks compared with placebo (mean change from baseline heart rate: +4.5 bpm with sibutramine v –1.4 bpm with placebo; P less than 0.0001).

We found two further RCTs assessing the effects of sibutramine on heart-valve function. Both of these RCTs may have been too small to detect clinically important adverse effects. The first RCT (210 obese people) compared sibutramine versus placebo for 12 months. It found no significant difference in the incidence of valvular disease between sibutramine and placebo (3/133 [2.3%] with sibutramine 15–20 mg/day v 2/77 [2.6%] with placebo; OR 0.87, 90% CI 0.19 to 3.97). The trial gave no information on efficacy. The second RCT (184 obese people) compared sibutramine 10 mg or 20 mg daily versus placebo. It reported no change in valvular appearance on echocardiogram in any group (no statistical comparisons between or within groups reported).

Sibutramine versus orlistat:

The systematic review gave no information on treatment-related adverse effects. The review found no significant difference between sibutramine and orlistat in rate of withdrawal (4 RCTs, 369 people; 9/184 [5%] with sibutramine v 16/185 [9%] with orlistat; RR 0.6, 95% CI 0.3 to 1.4). However, the review noted that three of four RCTs included in the meta-analysis reported a lower risk of withdrawal with sibutramine compared with orlistat.

Sibutramine versus diethylpropion, mazindol, rimonabant, or phentermine:

We found no RCTs comparing sibutramine versus other drugs.

Sibutramine plus orlistat:

See harms of sibutramine plus orlistat.

Drug safety alert

The European Medicines Agency (EMEA) has suspended marketing of sibutramine because of the increased risk of non-fatal myocardial infarctions and strokes (21 January 2010).

The European Medicines Agency (EMEA) has suspended marketing of sibutramine because of the increased risk of non-fatal myocardial infarctions and strokes. (www.ema.europa.eu)

Comment

Clinical guide:

Sibutramine has been associated with increases in systolic and diastolic blood pressure, heart rate, and total as well as LDL cholesterol; conversely, it has been associated with modest decreases in triglyceride levels, fasting serum glucose levels, glycosylated haemoglobin levels, and modest increases in HDL cholesterol levels.

Substantive changes

Sibutramine One systematic review and two subsequent RCTs added support evidence previously reported on the effects of sibutramine on weight loss compared with placebo. All three reports found that sibutramine increased weight loss at 4 to 12 months compared with placebo. The review and one RCT also reported that, compared with placebo, sibutramine increased the proportion of people who lost 5% or more of their initial body weight. One systematic review added comparing sibutramine versus orlistat found that sibutramine increased weight loss compared with orlistat, but it found no significant difference between treatments in people with type 2 diabetes or hypertension.In January 2010, the European Medicines Agency (EMEA) suspended marketing authorisation of sibutramine in the European Union because of the increased risk of non-fatal myocardial infarctions and strokes. Categorisation unchanged (trade-off-between-benefits-and-harms).

BMJ Clin Evid. 2010 Feb 11;2010:0604.

Sibutramine plus orlistat

Summary

WEIGHT LOSS Compared with orlistat alone: Sibutramine plus orlistat seems more effective at increasing weight loss at 6 months in obese people who are also on a reduced-calorie diet ( moderate-quality evidence ). Compared with sibutramine alone: We don't know whether sibutramine plus orlistat is more effective at increasing weight loss at 6 months in obese people who are also on a reduced-calorie diet or at maintaining weight loss at 16 weeks in obese people with a modified lifestyle who have taken it for 1 year ( very low-quality evidence ). NOTE We found no direct information from RCTs about whether sibutramine plus orlistat is better than no active treatment.

Benefits

Sibutramine plus orlistat versus placebo:

We found no systematic review or RCTs.

Sibutramine plus orlistat versus orlistat alone:

We found one RCT comparing three treatments for weight loss for 6 months: sibutramine (15 mg daily) plus orlistat (120 mg 3 times daily), sibutramine alone (15 mg daily), and orlistat alone (120 mg 3 times daily). All people enrolled also followed a reduced-calorie diet. It found that sibutramine plus orlistat significantly increased mean weight loss compared with orlistat alone at 6 months (3-arm RCT; 89 obese women; mean weight loss: 10.8 kg with sibutramine plus orlistat v 5.5 kg with orlistat alone; P = 0.002).

Sibutramine plus orlistat versus sibutramine alone:

We found two RCTs, one assessing weight loss and the other assessing weight-loss maintenance. The first RCT compared three treatments for weight loss for 6 months: sibutramine (15 mg daily) plus orlistat (120 mg 3 times daily), sibutramine alone (15 mg daily), and orlistat alone (120 mg 3 times daily). All people enrolled also followed a reduced-calorie diet. The RCT found no significant difference between sibutramine plus orlistat and sibutramine alone in weight loss at 6 months (3-arm RCT; 89 obese women; mean weight loss: 10.8 kg with sibutramine plus orlistat v 10.1 kg with sibutramine alone; P = 0.35). The second RCT compared sibutramine (10–15 mg daily) plus orlistat (120 mg 3 times daily) versus sibutramine plus placebo for weight-loss maintenance. It found no significant difference between treatments in mean body weight after 16 weeks (34 women who had completed 1 year of sibutramine alone plus lifestyle modification; mean change in body weight: +0.1 kg with sibutramine plus orlistat v +0.5 kg with sibutramine plus placebo; difference reported as not significant; P value not reported). Considering the small study size and the large withdrawal rate (only 76% of the women completed the study), these results should be interpreted with caution.

Sibutramine plus orlistat versus other drugs:

We found no RCTs.

Harms

Sibutramine plus orlistat versus placebo:

We found no RCTs.

Sibutramine plus orlistat versus orlistat alone:

The RCT gave no information on adverse effects.

Sibutramine plus orlistat versus sibutramine alone:

The first RCT gave no information on adverse effects. The second RCT found that sibutramine plus orlistat significantly increased GI adverse effects (soft stools, bowel movements, oily evacuation, and faecal urge) compared with sibutramine plus placebo at 68 weeks (soft stools: 50% with sibutramine plus orlistat v 9% with sibutramine plus placebo; P = 0.04; increased frequency of bowel movements: 50% with sibutramine plus orlistat v 9% with sibutramine plus placebo; P = 0.04; oily evacuation: 43% with sibutramine plus orlistat v 0% with sibutramine plus placebo; P = 0.02; faecal urgency: 43% with sibutramine plus orlistat v 9% with sibutramine plus placebo; P = 0.02).

For further information on adverse effects associated with sibutramine, see harms of sibutramine alone.

Sibutramine plus orlistat versus other drugs:

We found no RCTs.

Drug safety alert

The European Medicines Agency (EMEA) has suspended marketing of sibutramine because of the increased risk of non-fatal myocardial infarctions and strokes (21 January 2010).

The European Medicines Agency (EMEA) has suspended marketing of sibutramine because of the increased risk of non-fatal myocardial infarctions and strokes. (www.ema.europa.eu)

Comment

None.

Substantive changes

No new evidence

BMJ Clin Evid. 2010 Feb 11;2010:0604.

Diethylpropion

Summary

Compared with phentermine: Diethylpropion may be less effective at increasing weight loss ( low-quality evidence ). NOTE We found no direct information from RCTs about whether diethylpropion is better than no active treatment in the treatment of adults with obesity.

Benefits

Diethylpropion versus placebo:

We found no systematic review or RCTs.

Diethylpropion versus phentermine:

See benefits of phentermine.

Diethylpropion versus orlistat:

See benefits of orlistat.

Diethylpropion versus sibutramine:

See benefits of sibutramine.

Harms

Diethylpropion versus placebo:

We found no RCTs.

Diethylpropion versus phentermine:

See harms of phentermine.

Diethylpropion versus orlistat:

See harms of orlistat.

Diethylpropion versus sibutramine:

See harms of sibutramine.

Comment

None.

Substantive changes

Diethylpropion New option for which we found one systematic review. The review found less weight loss with diethylpropion compared with phentermine. Data were from one small RCT and effects of diethylpropion in weight loss are unclear. Categorised as Unknown effectiveness.

BMJ Clin Evid. 2010 Feb 11;2010:0604.

Mazindol

Summary

NOTE We found no direct information from RCTs about whether mazindol is better than no active treatment or other drugs in the treatment of adults with obesity.

Benefits

Mazindol versus placebo:

We found no systematic review or RCTs.

Mazindol versus orlistat:

See benefits of orlistat.

Mazindol versus sibutramine:

See benefits of sibutramine.

Harms

Mazindol versus placebo:

We found no RCTs.

Mazindol versus orlistat:

See harms of orlistat.

Mazindol versus sibutramine:

See harms of sibutramine.

Comment

None.

Substantive changes

Mazindol New option for which we found no evidence. Categorised as Unknown effectiveness.

BMJ Clin Evid. 2010 Feb 11;2010:0604.

Bariatric surgery versus non-surgical treatment

Summary

WEIGHT LOSS Compared with non-surgical treatment: Bariatric surgery (gastric banding, gastric bypass, vertical banded gastroplasty, and biliopancreatic diversion) may be more effective than non-surgical treatments (predominantly lifestyle modification to reduce calorie intake and increase exercise) at increasing weight loss and percentage of initial weight lost at 1 to 10 years ( low-quality evidence ). NOTE We found no direct information from RCTs on the effects of sleeve gastrectomy compared with non-surgical treatment.

Benefits

We found one systematic review (search date 2008), which identified two RCTs and three cohort studies in people with morbid obesity. The RCTs and cohorts identified compared various bariatric surgical techniques (gastric banding, gastric bypass, vertical banded gastroplasty, and biliopancreatic diversion) versus non-surgical treatment. Types of surgery differed between RCTs and cohort studies, and some studies combined data for different surgical techniques. There was also considerable variation in the non-surgical treatment given.

The first RCT (60 people; BMI 30–40 kg/m2 [mean BMI of more than 37.0 kg/m2 in both groups]; all with type 2 diabetes) identified by the review compared laparoscopic adjustable gastric banding versus non-surgical treatment (lifestyle modification programme structured to reduce calorie intake and increase exercise; drug treatments also allowed). At 2 years, the review found that, compared with non-surgical treatment, laparoscopic adjustable gastric banding significantly increased mean weight loss and percentage of initial weight lost (mean weight loss: –21.1 kg with laparoscopic adjustable gastric banding v –1.5 kg with non-surgical treatment; difference –19.6 kg, 95% CI –23.8 kg to –15.2 kg; P less than 0.001; percentage of initial weight lost: 20.0% with laparoscopic adjustable gastric banding v 1.4% with non-surgical treatment; P less than 0.001).

The second RCT (79 people; BMI more than 40 kg/m2) identified by the review compared biliopancreatic diversion versus non-surgical treatment (calorie-controlled diet; modified every 6 months). The review found that weight loss at 1 year was greater with biliopancreatic diversion in men and women compared with calorie-controlled diet (change in weight from baseline; women: from 125.3 kg to 90.2 kg with biliopancreatic diversion v from 121.6 kg to 114.5 kg with calorie-controlled diet; men: from 151.8 kg to 99.7 kg with biliopancreatic diversion v from 147.3 kg to 138.2 kg with calorie-controlled diet; significance not assessed).

The first cohort study (4047 people: mean BMI of 42.4 kg/m2 in surgical group and 40.1 kg/m2 in non-surgical group) identified by the review was a multicentre, prospective cohort study comparing bariatric surgery versus usual care. Eligible people self-selected either bariatric surgery or non-surgical (usual) care. Each person who selected surgical treatment was matched on 18 clinical variables with a person from the usual care group. Each surgeon determined the surgical procedure offered: vertical banded gastroplasty (more than 70%), gastric bypass (6%), or gastric banding (23%). Usual care was according to local practice and usually did not include pharmacotherapy. The study found that bariatric surgery significantly increased weight loss at 2 years compared with usual care (3505 people assessed; mean change in body weight: –23.4% with surgery v +0.1% with usual care; P less than 0.001). Long-term follow-up found that the difference in weight loss between groups remained significant at 10 years (1703 people assessed; mean change in body weight: –16.1% with surgery v +1.6% with usual care; P less than 0.001). There was variation in the mean percentage of weight lost at 10 years with type of bariatric procedure performed (mean weight loss [number of people assessed]: –25.0% with gastric bypass [58/265 people] v –16.0% with vertical banded gastroplasty [746/1369 people] v –14.0% with gastric banding [237/376 people]; significance not assessed).

The second cohort study (93 people [mean BMI more than 40 kg/m2 in both groups], with 63 people in the surgery group and 30 people in the non-surgical treatment group) identified by the review compared laparoscopic adjustable gastric banding or laparoscopic gastric bypass versus non-surgical treatment. The review found that mean excess-weight loss at a mean follow-up of 3.2 years was significantly greater with surgery compared with non-surgical treatment (42.2% with surgery v 11.5% with non-surgical treatment; P less than 0.001).

The third cohort study (20 people; mean BMI more than 40 kg/m2 in both groups) identified by the review compared laparoscopic adjustable gastric banding versus gastric bypass versus non-surgical treatment. The review found that mean BMI at 2 years was significantly lower with both surgeries compared with non-surgical treatment (32.9 kg/m2 with gastric bypass v 33.2 kg/m2 with laparoscopic adjustable gastric banding v 41.0 kg/m2 with non-surgical treatment; P less than 0.001 for both comparisons of surgery versus non-surgical treatment).

Harms

The RCT comparing laparoscopic adjustable gastric banding versus non-surgical treatment reported no deaths in either arm. The second RCT gave no information on adverse effects.

The first cohort study reported five postoperative deaths in 2010 people (0.25%): three deaths owing to leakage; one owing to a technical mistake during laparoscopic surgery; and one owing to postoperative MI. It also reported that 13% (151/1164) of people experienced postoperative complications (including bleeding: 0.9%; thromboembolism: 0.8%; wound complications: 1.8%; deep infections: 2.1%; pulmonary complications: 6.1%; and other complications: 4.8%; absolute numbers not reported), and that 26/1164 (2%) people required re-operation.

The second cohort study gave information on only re-operation rate. It reported that 7/56 (13%) of people who underwent laparoscopic gastric banding required re-operation. The cohort study also reported that 24% (9/37) people in the non-surgical group underwent gastric bypass.

The third cohort study gave no information on adverse effects.

Comment

Clinical guide:

Evidence regarding the efficacy and safety of bariatric surgical procedures comes from studies of mostly young, white women. Therefore, these results may not be generalisable to other populations.

Substantive changes

Bariatric surgery New option for which we found one systematic review. The review identified two RCTs and three cohort studies assessing the effects of various bariatric surgical techniques (gastric banding, gastric bypass, vertical banded gastroplasty, biliopancreatic diversion, and sleeve gastrectomy) in people with morbid obesity. All studies identified found that, compared with non-surgical treatment, bariatric surgery reduced weight loss and percentage of initial weight lost. Although operative complications are common, it is thought that the benefits of surgery outweigh the associated risks. Categorised as Likely to be beneficial.

BMJ Clin Evid. 2010 Feb 11;2010:0604.

Gastric bypass

Summary

WEIGHT LOSS Compared with vertical banded gastroplasty: Gastric bypass seems more effective at increasing weight loss at 1 to 3 years ( moderate-quality evidence ). Compared with gastric banding: Gastric bypass seems more effective at increasing weight loss at 5 years (moderate-quality evidence). Compared with sleeve gastrectomy: Gastric bypass seems less effective at increasing excess-weight loss at 1 year compared with sleeve gastrectomy (moderate-quality evidence). Proximal compared with distal gastric bypass: Proximal and distal gastric bypass seem equally effective at increasing weight loss at 1 to 3 years (moderate-quality evidence). Open gastric bypass compared with laparoscopic gastric bypass: We don't know whether open gastric bypass is more effective than laparoscopic gastric bypass at increasing weight loss ( low-quality evidence ). ADVERSE EFFECTS Gastric bypass is associated with nutritional and electrolyte abnormalities, GI symptoms, and surgical complications. There is a small risk of perioperative death with gastric bypass, but postoperative complications are common and may require re-operation. Laparoscopic gastric bypass reduces hospital stay compared with open gastric bypass. NOTE We found no clinically important results from RCTs about gastric bypass compared with biliopancreatic diversion in obese people.

Benefits

We found one systematic review (search date 2003) reporting that gastric bypass resulted in a mean weight loss of 43.5 kg (95% CI 41.2 kg to 43.5 kg; 32 RCTs, non-randomised controlled trials, and case series; 2937 people) at 12-month follow-up, and 41.5 kg (95% CI 37.4 kg to 45.6 kg; 21 RCTs, non-randomised controlled trials, and case series; 1281 people) at 36-month follow-up and beyond.

Gastric bypass versus vertical banded gastroplasty:

We found one systematic review (search date 2008; 7 RCTs, 435 people), which identified six RCTs comparing gastric bypass versus vertical banded gastroplasty, and one RCT comparing three interventions. The systematic review did not pool data and so we report data from the individual RCTs. Two of the RCTs were reported as only abstracts and are not discussed further. Data reported are as reported in the review.

The first RCT (42 adults with BMI 40 kg/m2 or more) found that gastric bypass significantly increased weight loss at 12 months compared with vertical banded gastroplasty (% excess weight lost: 78% with gastric bypass v 52% with vertical banded gastroplasty; P less than 0.05; data estimated by review from figure).

The second RCT (40 adults who were more than 44 kg overweight) also found that gastric bypass significantly increased weight loss compared with vertical banded gastroplasty at 12 months, 2 years, and 3 years (% of excess weight lost: 12 months: 68% with gastric bypass v 43% with vertical banded gastroplasty; P less than 0.001; 2 years: 66% with gastric bypass v 39% with vertical banded gastroplasty; P less than 0.001; 3 years: 62% with gastric bypass v 37% with vertical banded gastroplasty; P less than 0.001; data for 2 and 3 years estimated by review from a figure).

The third RCT (80 people with morbid obesity) found that laparoscopic gastric bypass significantly increased weight loss compared with laparoscopic vertical banded gastroplasty at 1 year (% excess weight lost: 63% with gastric bypass v 55% with vertical banded gastroplasty; reported as significant; P value not reported) and 2 years' follow-up (% excess weight lost: 71% with gastric bypass v 53% with vertical banded gastroplasty; reported as significant; P value not reported).

The fourth RCT (83 people with morbid obesity) also found that laparoscopic gastric bypass significantly increased weight loss compared with laparoscopic vertical banded gastroplasty at 1 and 2 years' follow-up (% excess weight lost; at 1 year: 78% with gastric bypass v 63% with vertical banded gastroplasty; P = 0.009; at 2 years: 84% with gastric bypass v 60% with vertical banded gastroplasty; P less than 0.001).

The fifth RCT (106 people; mean BMI of more than 48 kg/m2 in both groups) found no significant difference in success rate (defined as proportion of people with BMI of less than 35 kg/m2 or less than 50% excess weight) between the two procedures at 3 years and 5 to 6 years (at 3 years: 30/52 [58%] with gastric bypass v 21/54 [39%] with vertical banded gastroplasty; P = 0.08; at 5 to 6 years: 16/52 [34%] with gastric bypass v 9/54 [16%] with vertical banded gastroplasty; P = 0.112).

Gastric bypass versus gastric banding:

We found one systematic review (search date 2008), which identified one RCT. The RCT (51 adults with BMI between 35 kg/m2 and 50 kg/m2 [mean BMI more than 43.0  kg/m2 in both groups]) compared laparoscopic gastric bypass versus laparoscopic adjustable gastric banding. It found that gastric bypass significantly increased excess-weight loss at 5 years (% excess weight lost: 66.6% with gastric bypass v 47.5% with gastric banding; P less than 0.001).

Gastric bypass versus biliopancreatic diversion:

We found one systematic review (search date 2008), which identified no RCTs comparing gastric bypass versus biliopancreatic diversion.

Gastric bypass versus sleeve gastrectomy:

We found one systematic review (search date 2008), which identified one small RCT. The RCT (32 people with BMI more than 35 kg/m2 [mean BMI more than 45.0  kg/m 2in both groups]) found that sleeve gastrectomy significantly increased excess-weight loss at 1 year compared with gastric bypass (% excess weight lost: 60.5% with gastric bypass v 69.7% with sleeve gastrectomy; P = 0.05).

Proximal versus distal gastric bypass:

We found three RCTs comparing proximal versus distal laparoscopic gastric bypass.

The first RCT (48 people with mean BMI 44.9 kg/m2) found no significant difference in mean weight or mean excess-weight loss between proximal and distal gastric bypass at up to 2 years' follow-up (mean weight at 1 year: 76.3 kg with proximal v 79.0 kg with distal; mean weight at 2 years: 90 kg with proximal v 83.2 kg with distal; mean excess weight lost: results displayed graphically; differences reported as not significant; P values not reported).

The second RCT (133 people with BMI more than 40 kg/m2) found no significant difference in mean weight loss, change in BMI, or percentage of excess-weight loss over time between proximal and distal gastric bypass at up to 3 years of follow-up (reported as not significant; absolute results presented graphically). A larger proportion of people with a BMI more than 50 kg/m2 in the distal gastric bypass group had greater success — defined by number achieving loss of more than 50% excess weight — at 18 months but not at 24 or 36 months compared with the proximal gastric bypass group (results displayed graphically; differences reported as not significant; P value not reported).

The third RCT (105 people with BMI 50 kg/m2 or more) found no significant difference between proximal and distal gastric bypass in excess-weight loss at 4 years (mean % excess weight lost: 70% with proximal v 74% with distal; reported as not significant; P value not reported).

Open versus laparoscopic gastric bypass:

We found one systematic review (search date 2008; 4 RCTs, 360 people) comparing open versus laparoscopic gastric bypass. The review did not pool data and so we report data from the individual RCTs identified: data reported are as reported in the review. One RCT was reported as only an abstract and is not discussed further.

The first RCT (155 people with BMI 40–60 kg/m2) identified by the review found no significant difference between open and laparoscopic techniques in excess-weight loss at 3 years (% excess weight lost: 77% with laparoscopic gastric bypass v 67% with open gastric bypass; reported as non-significant; P value not reported).

The second RCT (50 people with BMI more than 40 kg/m2) identified by the review found that weight loss at 1 year was similar in each group (weight loss: 41 kg with open v 39 kg with laparoscopic; significance not assessed).

The third RCT (104 people with morbid obesity) identified by the review found no significant difference in weight loss at a mean follow-up of 23 months between open and laparoscopic techniques (weight loss displayed graphically; between-group difference reported as not significant; P value not reported).

Harms

One systematic review (search date 2003) found that gastric bypass resulted in a 30-day mortality rate of 1.0% in comparative studies (95% CI 0.5% to 1.9%; 15 RCTs and non-randomised controlled trials, 907 people) and 0.3% in case series (95% CI 0.2% to 0.4%; 50 case series, 11,290 people). Adverse effects were common, and included surgical complications (19%), nutritional and electrolyte abnormalities (17%), and GI symptoms (17%).

Gastric bypass versus vertical banded gastroplasty:

The review did not pool data on adverse effects of surgeries. Four RCTs identified by the review comparing gastric bypass versus vertical banded gastroplasty reported no deaths. One RCT identified by the review reported no deaths with vertical banded gastroplasty, but two deaths (10%) with gastric bypass, which occurred after 3 days and 12 months, were caused by presumed arrhythmia.

One RCT identified by the review found that gastric bypass significantly increased the rate of early postoperative complications (including anastomotic leakage, abdominal abscess, and GI bleeding) compared with vertical banded gastroplasty (7/40 [18%] with gastric bypass v 1/40 [3%] with vertical banded gastroplasty; P less than 0.05). Another RCT identified by the review found no significant difference between interventions in re-operation rates or perioperative complications (re-operation: 5/37 [14%] with gastric bypass v 1/46 [2%] with vertical banded gastroplasty; P = 0.08; minor bleeding: 2/37 [5%] with gastric bypass v 4/46 [9%] with vertical banded gastroplasty; suspected leakage: 1/37 [2.7%] with gastric bypass v 1/46 [2.2%] with vertical banded gastroplasty; difference reported as not significant for all outcomes; P values not reported for minor bleeding or suspected leakage).

Gastric bypass versus gastric banding:

The RCT identified by the review reported no deaths. The RCT found similar rates of re-operation for each technique (3/24 [13%] with gastric bypass v 4/26 [15%] with gastric banding; significance not assessed). People in the gastric banding group underwent re-operation for inadequate weight loss or pouch dilation. Reasons were not specified for re-operations in the gastric bypass group. Early and late complication rates were similar among procedures.

Gastric bypass versus biliopancreatic diversion:

We found no RCTs.

Gastric bypass versus sleeve gastrectomy:

The RCT identified by the review reported no deaths and no intraoperative or postoperative complications. No other details were reported.

Proximal versus distal gastric bypass:

The first RCT reported no deaths in either group. It found no significant difference in overall complication rates between proximal and distal gastric bypass (figures not reported; reported as not significant; P value not reported). However, the incidence of internal hernias was significantly lower with proximal bypass compared with distal bypass (0/25 [0%] with proximal bypass v 4/23 [17%] with distal bypass; P = 0.029). The second RCT reported one death in the proximal gastric bypass group and one death in the distal gastric bypass group within the first 30 days. It gave no other information on complications. The third RCT reported no deaths in either group. The RCT reported that rates of re-operation were similar in each group (2/57 [4%] with proximal bypass v 1/48 [2%] with distal bypass; significance not assessed).

Open versus laparoscopic gastric bypass:

The RCTs identified by the review reported four postoperative deaths: one caused by malignant hyperthermia; one caused by possible pulmonary thromboembolism (laparoscopic); one caused by intestinal obstruction (laparoscopic); and one caused by evisceration. The first RCT found no significant difference between open and laparoscopic bypass in the proportion of people who had major surgical complications (9% with open gastric bypass v 8% with laparoscopic gastric bypass; absolute numbers not reported; P = 0.78). However, this RCT found that open gastric bypass was associated with a significantly higher rate of incisional hernia (22/57 [39%] with open gastric bypass v 3/59 [5%] with laparoscopic gastric bypass; P less than 0.01). In all three RCTs identified by the review, minor complications (including vomiting, colicky pain, and wound infection) were not significantly different between groups.

The second RCT found that open gastric bypass was associated with a significantly higher rate of late complications (including eventrations, abscess, intestinal obstruction, and pancreatitis) compared with laparoscopic gastric bypass (12/51 [24%] with open gastric bypass v 6/53 [11%] with laparoscopic gastric bypass; P less than 0.05). Operating time was significantly longer for the laparoscopic procedure in two of the RCTs (first RCT: 195 minutes with open v 225 minutes with laparoscopic; P less than 0.001; second RCT: 85 minutes with open v 150 minutes with laparoscopic), but it was significantly longer for the open procedure in one RCT (201 minutes with open v 186 minutes with laparoscopic: P less than 0.05). Hospital stay was significantly shorter for the laparoscopic procedure (4–8 days with open v 3–5 days with laparoscopic; P less than 0.05 in 2 RCTs).

Reinforcement of staple lines in gastric bypass versus non-reinforcement:

We found two RCTs that evaluated reinforcement of staple lines in gastric bypass surgery. The first RCT (34 people with BMI of 40–60 kg/m2) evaluated glycolide copolymer staple-line reinforcement sleeves versus non-reinforcement during laparoscopic gastric bypass. The RCT found that reinforcement of the staple lines was associated with significantly fewer bleeding sites during construction of the gastric pouch, the division of the jejunum, and the division of the jejunal mesenteric tissue (gastric pouch: 2.5 without reinforcement v 0.4 with reinforcement; division of the jejunum: 0.6 without reinforcement v 0.2 with reinforcement; division of the jejunal mesenteric tissue: 0.8 without reinforcement v 0 with reinforcement; P less than 0.01 for all comparisons). One person in the non-reinforcement arm had postoperative GI haemorrhage requiring transfusion and re-operation.

The second RCT (48 people with BMI more than 35 kg/m2) evaluated polyglycolide acid and trimethylene carbonate staple-line reinforcement versus non-reinforcement during laparoscopic gastric bypass. The RCT found that, compared with non-reinforcement, reinforcement of staple lines was associated with a significantly shorter operation time (average time: 115 minutes with reinforcement v 150 minutes without reinforcement; P = 0.03), and significantly higher postoperative haemoglobin levels (mean haemoglobin level: 12.5 g/dL with reinforcement v 11.1 g/dL without reinforcement; P = 0.0156). No patients required re-operation or transfusion in either group.

Comment

None.

Substantive changes

Gastric bypass One updated review identified no new evidence on the effects of gastric bypass versus vertical banded gastroplasty. One small RCT identified by the review found that sleeve gastrectomy increased excess-weight loss at 1 year compared with gastric bypass. One small RCT identified by the review found that gastric bypass increased excess-weight loss at 5 years compared with gastric banding. One RCT added comparing proximal versus distal gastric bypass found no significant difference between the two techniques in mean excess-weight loss at 4 years. One RCT added on the effects of staple reinforcement found that, compared with non-reinforcement, reinforcement of staple lines was associated with a shorter operation time and higher postoperative haemoglobin levels. Categorisation unchanged (Likely to be beneficial).

BMJ Clin Evid. 2010 Feb 11;2010:0604.

Gastric banding

Summary

WEIGHT LOSS Compared with gastric bypass: Gastric banding seems less effective at increasing weight loss at 5 years ( moderate-quality evidence ). Compared with vertical banded gastroplasty: We don't know whether gastric banding is more effective at increasing weight loss at 5 years ( very low-quality evidence ). Compared with sleeve gastrectomy: Gastric banding may be less effective at increasing weight loss at 1 and 3 years ( low-quality evidence ). Open versus laparoscopic gastric banding: We don't know whether open or laparoscopic gastric banding is more effective at increasing weight loss at 12 months (low-quality evidence). NOTE We found no clinically important results from RCTs about gastric banding compared with biliopancreatic diversion in people with obesity.

Benefits

One systematic review (search date 2003) found that gastric banding resulted in a mean weight loss of 30.2 kg (95% CI 28.0 kg to 32.4 kg; 27 RCTs, non-randomised controlled trials and case series; 5562 people) at 12-month follow-up and 34.8 kg (95% CI 29.5 kg to 40.1 kg; 17 RCTs, non-randomised controlled trials and case series; 3076 people) at 36-month follow-up and beyond.

Gastric banding versus gastric bypass:

See benefits of gastric bypass.

Gastric banding versus vertical banded gastroplasty:

We found one systematic review (search date 2008; 3 RCTs, 259 people). The review did not pool data and so we report data separately from the individual RCTs. Data reported are as reported in the review.

The first RCT (59 adults with BMI 40 kg/m2 or more or BMI 37 kg/m2 or more with associated comorbidity) identified by the review found that the gastric banding group had lost less weight at 1 year compared with the vertical banded gastroplasty group (change in mean weight from baseline: from 124 kg to 98 kg with gastric banding v from 123 kg to 82 kg with vertical banded gastroplasty; significance not assessed). However, at 5 years, the gastric banding group had lost more weight (from 124 kg to 81 kg with gastric banding v from 123 kg to 88 kg with vertical banded gastroplasty; significance not assessed).

The second RCT (100 adults with BMI 40–50 kg/m2) found that a significantly smaller proportion of people achieved an excellent or good result (defined as residual excess weight of less than 50%) at 2 years in the laparoscopic gastric banding group compared with the laparoscopic vertical banded gastroplasty group (35% with gastric banding v 74% with vertical banded gastroplasty; absolute numbers not reported; P less than 0.001). There was no significant difference between groups in excellent or good result rates at 3 years, although the proportion of people classed as having an excellent or good result was smaller with gastric banding (25% with gastric banding v 63% with vertical banded gastroplasty; P = 0.056). The RCT also found no significant difference between the two surgeries in excess-weight loss at 2 years (% excess weight lost: 41% with gastric banding v 64% with vertical banded gastroplasty; reported as not significant; P value not reported) and at 3 years (% excess weight lost: 39% with gastric banding v 59% with vertical banded gastroplasty; reported as not significant; P value not reported).

The third RCT (100 adults with BMI more than 40 kg/m2 or more than 35 kg/m2 with associated comorbidity) found that laparoscopic gastric banding was significantly less effective at reducing excess weight at 1 and 2 years compared with open vertical banded gastroplasty (% excess weight lost: at 1 year: 53.3% with gastric banding v 71.1% with vertical banded gastroplasty; 2 years: 54.9% with gastric banding v 70.1% with vertical banded gastroplasty; P = 0.001 or less at both time points).

Gastric banding versus sleeve gastrectomy:

We found one systematic review (search date 2008), which identified one RCT (80 adults with BMI more than 30 kg/m2 [median BMI of 37 kg/m2 in the gastric banding group and 39 kg/m2 in the sleeve gastrectomy group]) comparing laparoscopic gastric banding versus laparoscopic isolated sleeve gastrectomy.

The RCT found that weight loss and excess-weight loss at 1 and 3 years were significantly less with gastric banding compared with sleeve gastrectomy (median weight loss at 1 year: –14 kg with gastric banding v –26 kg sleeve gastrectomy; P less than 0.0001; median weight loss at 3 years: –17 kg with gastric banding v –29.5 kg sleeve gastrectomy; P less than 0.0001; % excess weight lost at 1 year: 41.4% with gastric banding v 57.7% with sleeve gastrectomy; P = 0.0004; % excess weight lost at 3 years: 48% with gastric banding v 66% with sleeve gastrectomy; P = 0.0025).

Open versus laparoscopic gastric banding:

We found one systematic review (search date 2008), which identified one RCT (50 adults with BMI more than 40 kg/m2) comparing open versus laparoscopic gastric banding. The RCT found no significant difference in weight loss at 12 months between open and laparoscopic gastric banding (weight loss: –34.4 kg with open v –35.0 kg with laparoscopic gastric banding; reported as not significant; P value not reported).

Harms

One systematic review (search date 2003) found that gastric banding resulted in a 30-day mortality rate of 0.4% in comparative studies (95% CI 0.01% to 2.1%; 6 RCTs and non-randomised controlled trials; 268 people) and 0.02% in case series (95% CI 0% to 0.78%; 35 case series, 9222 people). Adverse effects were common and included surgical complications (13.2%), re-operations (7.7%), and GI symptoms (7.0%).

Gastric banding versus gastric bypass:

See harms of gastric bypass.

Gastric banding versus vertical banded gastroplasty:

The review did not pool data on adverse effects.

The first RCT identified by the review reported one death from each group during the follow-up period, but neither death was attributed to the surgery. Re-operations occurred in 10/27 (37%) people having vertical banded gastroplasty and 3/26 (12%) people having gastric banding. Gastro-oesophageal reflux was more common in people having vertical banded gastroplasty compared with people having gastric banding (4/27 [15%] with gastroplasty v 3/26 [12%] with gastric banding; significance not assessed).

The second RCT identified by the review reported no deaths in either arm. The RCT found that gastric banding significantly increased the proportion of people who required re-operation compared with vertical banded gastroplasty (12/49 [25%] with gastric banding v 0/50 [0%] with vertical banded gastroplasty; P less than 0.05). It also found that gastric banding significantly increased late complications, such as pouch dilatation, pouch to fundus fistula, symptomatic reflux disease, and gastric bezoar compared with vertical banded gastroplasty (16/49 [33%] with gastric banding v 7/50 [14%] with vertical banded gastroplasty; P less than 0.001).

The third RCT identified by the review reported two (2/50 [4%]) deaths in the vertical banded gastroplasty group, but none in the gastric banding group. It found that perioperative complications were more common in the vertical banded gastroplasty group than in the gastric banding group (3/50 [6%] with gastric banding v 9/50 [18%] with vertical banded gastroplasty; significance not assessed) and that rate of late complications requiring further surgery was similar in each group (20/50 [40%] with gastric banding v 26/50 [52%] with vertical banded gastroplasty; significance not assessed).

Gastric banding versus sleeve gastrectomy:

The RCT identified by the review reported no deaths in either arm. Two people who had sleeve gastrectomy required re-operation for early postoperative complications (intraperitoneal bleed and gastric ischaemia), but no one in the gastric band group required re-operation for early postoperative complications (significance not assessed for between group comparison). Late complications requiring surgery were more common in the gastric band group compared with the sleeve gastrectomy group (7/40 [18%] with gastric banding v 0/40 [0%] with sleeve gastrectomy; significance not assessed).

Open versus laparoscopic gastric banding:

The RCT identified by the review reported no deaths in either arm. It also found no significant difference in surgical complications between the two procedures (no further details given reported as not significant; P value not reported). It found no significant difference between open gastric banding and laparoscopic gastric banding in rate of incisional hernia complications, but the rate was higher with open gastric banding (3/25 [12%] with open v 0/25 [0%] with laparoscopic gastric banding; reported as not significant; P value not reported). It also found that open laparoscopic gastric banding was associated with a significantly higher rate of readmission at 1 year and significantly increased duration of hospital stay in the first year compared with laparoscopic gastric banding (readmissions: 15/25 [60%] with open v 6/25 [24%] with laparoscopic gastric banding; P less than 0.05; mean days in hospital: 11.8 days with open v 7.8 days with laparoscopic gastric banding; P less than 0.05).

Comment

None.

Substantive changes

Gastric banding New option for which we found one systematic review. The review assessed the effects of gastric banding versus various other bariatric surgeries. RCTs identified by the review found that gastric banding was less effective than gastric bypass, vertical banded gastroplasty, and sleeve gastrectomy at improving various weight-related outcomes at various lengths of follow-up. One RCT found no significant difference between open and laparoscopic gastric banding in weight loss at 12 months. Categorised as Likely to be beneficial.

BMJ Clin Evid. 2010 Feb 11;2010:0604.

Vertical banded gastroplasty

Summary

WEIGHT LOSS Compared with gastric banding: We don't know whether vertical banded gastroplasty is more effective at increasing weight loss at 5 years ( very low-quality evidence ). Compared with gastric bypass: Vertical banded gastroplasty seems less effective at increasing weight loss at 1 to 3 years ( moderate-quality evidence ). Open compared with laparoscopic vertical banded gastroplasty: We don't know whether open vertical banded gastroplasty is more effective at increasing weight loss at 12 months ( low-quality evidence ). ADVERSE EFFECTS There is a small risk of perioperative death with vertical banded gastroplasty, but GI symptoms and postoperative complications possibly requiring re-operation are common. Open vertical banded gastroplasty has a shorter operating time compared with laparoscopic banded gastroplasty. NOTE We found no clinically important results from RCTs about vertical banded gastroplasty compared with non-surgical treatment, or compared with biliopancreatic diversion or sleeve gastrectomy in obese people.

Benefits

We found one systematic review (search date 2003), which found that vertical banded gastroplasty resulted in a mean weight loss of 32.2 kg (95% CI 29.9 kg to 34.4 kg; 21 RCTs, non-randomised controlled trials, and case series; 2080 people; initial mean BMI not reported) at 12-month follow-up, and 32.0 kg (95% CI 27.7 kg to 36.4 kg; 18 RCTs, non-randomised controlled trials, and case series; 1877 people; initial mean BMI not reported) at 36-month follow-up and beyond.

Vertical banded gastroplasty versus gastric banding:

See benefits of gastric banding.

Vertical banded gastroplasty versus gastric bypass:

See benefits of gastric bypass.

Open versus laparoscopic vertical banded gastroplasty:

We found one systematic review (search date 2008), which identified one small RCT (30 adults with BMI 40–50 kg/m2). The RCT found similar weight loss between open and laparoscopic vertical banded gastroplasty at 12 months (mean weight loss: 55% with open v 47% with laparoscopic; significance not assessed).

Harms

One systematic review found that gastroplasty resulted in a 30-day mortality rate of 0.2% in controlled studies (95% CI 0% to 1.4%; 11 RCTs and non-randomised controlled trials; 401 people) and 0.3% in case series (95% CI 0.1% to 0.5%; 33 case series; 4091 people). Adverse effects were common, and included surgical complications (24%), re-operations (11%), and GI symptoms (18%).

Vertical banded gastroplasty versus gastric banding:

See harms of gastric banding.

Vertical banded gastroplasty versus gastric bypass:

See harms of gastric bypass.

Open versus laparoscopic vertical banded gastroplasty:

The RCT reported no deaths in either arm. Operating time was significantly longer with laparoscopic compared with open vertical banded gastroplasty (2.10 hours with laparoscopic v 1.45 hours with open; P less than 0.002). There was no significant difference between open and laparoscopic vertical banded gastroplasty in average hospital stay (4 days for both techniques; difference reported as not significant; P value not reported). Two people, one in each group, developed a fistula at the gastric partition that required re-operation. Two people having open gastroplasty developed abdominal-wall hernias at 12 months.

Comment

None.

Substantive changes

Vertical banded gastroplasty One updated systematic review found no new evidence on the effects of vertical banded gastroplasty. Categorisation unchanged (Likely to be beneficial).

BMJ Clin Evid. 2010 Feb 11;2010:0604.

Biliopancreatic diversion

Summary

NOTE We found no clinically important results from RCTs about biliopancreatic diversion compared with non-surgical treatment, or compared with gastric bypass, gastric banding, vertical banded gastroplasty, or sleeve gastrectomy in obese people.

Benefits

We found two systematic reviews (search dates 2003 and 2008), which found no RCTs comparing biliopancreatic diversion versus other bariatric surgery techniques. One of the reviews identified three case series of biliopancreatic diversion (735 people) that assessed weight loss. The review found that biliopancreatic diversion resulted in a mean weight loss of 51.9 kg (95% CI 45.1 kg to 58.8 kg; initial mean BMI not reported) at 12-month follow-up, and 53.1 kg (95% CI 47.4 kg to 58.8 kg; initial mean BMI not reported) at 36-month follow-up and beyond.

Biliopancreatic diversion versus gastric bypass:

See benefits of gastric bypass.

Harms

We found no RCTs. The systematic review (search date 2003) of case series found that biliopancreatic diversion resulted in a 30-day mortality rate of 0.9% (95% CI 0.5% to 1.3%; 7 case series; 2808 people). Adverse effects were common, and included surgical complications (6%), re-operations (4%), and GI symptoms (38%).

Biliopancreatic diversion versus gastric bypass:

See harms of gastric bypass.

Comment

None.

Substantive changes

Biliopancreatic diversion One updated systematic review found no new evidence on the effects of biliopancreatic diversion. Categorisation unchanged (Likely to be beneficial).

BMJ Clin Evid. 2010 Feb 11;2010:0604.

Sleeve gastrectomy

Summary

WEIGHT LOSS Compared with gastric banding: Sleeve gastrectomy may be more effective at increasing weight loss at 1 and 3 years ( low-quality evidence ). Compared with gastric bypass: Sleeve gastrectomy seems more effective at increasing mean excess-weight loss at 1 to 2 years ( moderate-quality evidence ). NOTE We found no clinically important results from RCTs about sleeve gastrectomy compared with non-surgical treatment, or compared with vertical banded gastroplasty or biliopancreatic diversion in obese people.

Benefits

Sleeve gastrectomy versus gastric banding:

See benefits of gastric banding.

Sleeve gastrectomy versus gastric bypass:

See benefits of gastric bypass.

Sleeve gastrectomy versus other bariatric surgery techniques:

We found two systematic reviews (search dates 2003 and 2008), which identified no RCTs comparing sleeve gastrectomy versus other bariatric surgery techniques.

Harms

Sleeve gastrectomy versus gastric banding:

See harms of gastric banding.

Sleeve gastrectomy versus gastric bypass:

See harms of gastric bypass.

Sleeve gastrectomy versus other bariatric surgery techniques:

We found no RCTs.

Comment

None.

Substantive changes

Sleeve gastrectomy One small RCT identified by an updated systematic review found that sleeve gastrectomy increased excess-weight loss at 1 year compared with gastric bypass. One small RCT identified by the review found that sleeve gastrectomy increased weight loss and excess-weight loss at 1 and 3 years compared with gastric banding. The RCTs were small and had methodological issues. Insufficient evidence to draw conclusions on the effectiveness of sleeve gastrectomy. Categorisation unchanged (Unknown effectiveness).

Articles from BMJ Clinical Evidence are provided here courtesy of BMJ Publishing Group

RESOURCES